feat: major architectural refactor

- Refactor memory system (episodic/STM/LTM with components)
- Implement complete subtitle domain (scanner, matcher, placer)
- Add YAML workflow infrastructure
- Externalize knowledge base (patterns, release groups)
- Add comprehensive testing suite
- Create manual testing CLIs

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.1.6"
+current_version = "0.1.7"
 parse = "(?P<major>\\d+)\\.(?P<minor>\\d+)\\.(?P<patch>\\d+)"
 serialize = ["{major}.{minor}.{patch}"]
 search = "{current_version}"

.dockerignore

@@ -22,8 +22,7 @@ venv
 .venv
 env
 .env
-.env.*
+.env-
-
 # IDE
 .vscode
 .idea
@@ -41,11 +40,8 @@ docs/
 *.md
 !README.md
 
-# Tests
+# Data
-tests/
+data/
-pytest.ini
-
-# Data (will be mounted as volumes)
 memory_data/
 logs/
 *.log

.env.alfred

@@ -0,0 +1,77 @@
+# --- IMPORTANT ---
+# Settings are split across multiple files for clarity.
+# Files (loaded in this order, last wins):
+#   .env.alfred   — app config and service addresses (safe to commit)
+#   .env.secrets  — generated secrets, passwords, URIs and API keys (DO NOT COMMIT)
+#   .env.make     — build metadata synced from pyproject.toml (safe to commit)
+#
+# To customize: edit .env.alfred for config, .env.secrets for secrets.
+
+# --- Alfred ---
+MAX_HISTORY_MESSAGES=10
+MAX_TOOL_ITERATIONS=10
+REQUEST_TIMEOUT=30
+
+# LLM Settings
+LLM_TEMPERATURE=0.2
+
+# Persistence
+DATA_STORAGE_DIR=data
+
+# Network
+HOST=0.0.0.0
+PORT=3080
+
+# --- DATABASES ---
+# Passwords and connection URIs are auto-generated in .env.secrets.
+# Edit host/port/user/dbname here if needed.
+
+# MongoDB (Application Data)
+MONGO_HOST=mongodb
+MONGO_PORT=27017
+MONGO_USER=alfred
+MONGO_DB_NAME=alfred
+
+# PostgreSQL (Vector Database / RAG)
+POSTGRES_HOST=vectordb
+POSTGRES_PORT=5432
+POSTGRES_USER=alfred
+POSTGRES_DB_NAME=alfred
+
+# --- EXTERNAL SERVICES ---
+
+# TMDB — Media metadata (required). Get your key at https://www.themoviedb.org/
+# → TMDB_API_KEY goes in .env.secrets
+TMDB_BASE_URL=https://api.themoviedb.org/3
+
+# qBittorrent
+# → QBITTORRENT_PASSWORD goes in .env.secrets
+QBITTORRENT_URL=http://qbittorrent:16140
+QBITTORRENT_USERNAME=admin
+QBITTORRENT_PORT=16140
+
+# Meilisearch
+# → MEILI_MASTER_KEY goes in .env.secrets
+# MEILI_ENABLED=false # KEY DOESN'T EXISTS => SEARCH IS THE PROPER KEY
+SEARCH=false
+MEILI_NO_ANALYTICS=true
+MEILI_HOST=http://meilisearch:7700
+
+# --- LLM CONFIGURATION ---
+# Providers: local, openai, anthropic, deepseek, google, kimi
+# → API keys go in .env.secrets
+DEFAULT_LLM_PROVIDER=local
+
+# Local LLM (Ollama)
+#OLLAMA_BASE_URL=http://ollama:11434
+#OLLAMA_MODEL=llama3.3:latest
+
+OLLAMA_BASE_URL=http://10.0.0.11:11434
+OLLAMA_MODEL=glm-4.7-flash:latest
+
+# --- RAG ENGINE ---
+RAG_ENABLED=TRUE
+RAG_API_URL=http://rag_api:8000
+RAG_API_PORT=8000
+EMBEDDINGS_PROVIDER=ollama
+EMBEDDINGS_MODEL=nomic-embed-text

.env.example

@@ -1,69 +1,73 @@
-# Agent Media - Environment Variables
+# --- IMPORTANT ---
+# Settings are split across multiple files for clarity.
+# Files (loaded in this order, last wins):
+#   .env.alfred   — app config and service addresses (safe to commit)
+#   .env.secrets  — generated secrets, passwords, URIs and API keys (DO NOT COMMIT)
+#   .env.make     — build metadata synced from pyproject.toml (safe to commit)
+#
+# To customize: edit .env.alfred for config, .env.secrets for secrets.
 
-# LibreChat Security Keys
+# --- Alfred ---
-# Generate secure keys with: openssl rand -base64 32
+MAX_HISTORY_MESSAGES=10
-JWT_SECRET=your-super-secret-jwt-key-change-this-in-production
+MAX_TOOL_ITERATIONS=10
-JWT_REFRESH_SECRET=your-super-secret-refresh-key-change-this-too
+REQUEST_TIMEOUT=30
 
-# Generate with: openssl rand -hex 16 (for CREDS_KEY)
+# LLM Settings
-CREDS_KEY=your-32-character-secret-key-here
+LLM_TEMPERATURE=0.2
 
-# Generate with: openssl rand -hex 8 (for CREDS_IV)
+# Persistence
-CREDS_IV=your-16-character-iv-here
+DATA_STORAGE_DIR=data
 
-# LibreChat Configuration
+# Network
-DOMAIN_CLIENT=http://localhost:3080
+HOST=0.0.0.0
-DOMAIN_SERVER=http://localhost:3080
+PORT=3080
 
-# Session expiry (in milliseconds)
+# --- DATABASES ---
-# Default: 15 minutes
+# Passwords and connection URIs are auto-generated in .env.secrets.
-SESSION_EXPIRY=900000
+# Edit host/port/user/dbname here if needed.
 
-# Refresh token expiry (in milliseconds)
+# MongoDB (Application Data)
-# Default: 7 days
+MONGO_HOST=mongodb
-REFRESH_TOKEN_EXPIRY=604800000
+MONGO_PORT=27017
+MONGO_USER=alfred
+MONGO_DB_NAME=LibreChat
 
-# Meilisearch Configuration
+# PostgreSQL (Vector Database / RAG)
-# Master key for Meilisearch (generate with: openssl rand -base64 32)
+POSTGRES_HOST=vectordb
-MEILI_MASTER_KEY=DrhYf7zENyR6AlUCKmnz0eYASOQdl6zxH7s7MKFSfFU
+POSTGRES_PORT=5432
+POSTGRES_USER=alfred
+POSTGRES_DB_NAME=alfred
 
-# PostgreSQL Configuration (for RAG API)
+# --- EXTERNAL SERVICES ---
-POSTGRES_DB=librechat_rag
-POSTGRES_USER=postgres
-POSTGRES_PASSWORD=postgres
 
-# RAG API Configuration (Vector Database)
+# TMDB — Media metadata (required). Get your key at https://www.themoviedb.org/
-RAG_COLLECTION_NAME=testcollection
+# → TMDB_API_KEY goes in .env.secrets
-RAG_EMBEDDINGS_PROVIDER=openai
+TMDB_BASE_URL=https://api.themoviedb.org/3
-RAG_EMBEDDINGS_MODEL=text-embedding-3-small
 
-# API Keys
+# qBittorrent
-# OpenAI API Key (required for RAG embeddings)
+# → QBITTORRENT_PASSWORD goes in .env.secrets
-OPENAI_API_KEY=your-openai-api-key-here
+QBITTORRENT_URL=http://qbittorrent:16140
-
-# Deepseek API Key (for LLM in agent-brain)
-DEEPSEEK_API_KEY=your-deepseek-api-key-here
-
-# Agent Brain Configuration
-
-# LLM Provider (deepseek or ollama)
-LLM_PROVIDER=deepseek
-
-# Memory storage directory (inside container)
-MEMORY_STORAGE_DIR=/data/memory
-
-# API Key for agent-brain (used by LibreChat custom endpoint)
-AGENT_BRAIN_API_KEY=agent-brain-secret-key
-
-# External Services (Optional)
-# TMDB API Key (for movie metadata)
-TMDB_API_KEY=your-tmdb-key
-
-# qBittorrent Configuration
-QBITTORRENT_URL=http://localhost:8080
 QBITTORRENT_USERNAME=admin
-QBITTORRENT_PASSWORD=adminpass
+QBITTORRENT_PORT=16140
 
-# Debug Options
+# Meilisearch
-DEBUG_LOGGING=false
+# → MEILI_MASTER_KEY goes in .env.secrets
-DEBUG_CONSOLE=false
+MEILI_ENABLED=FALSE
+MEILI_NO_ANALYTICS=TRUE
+MEILI_HOST=http://meilisearch:7700
+
+# --- LLM CONFIGURATION ---
+# Providers: local, openai, anthropic, deepseek, google, kimi
+# → API keys go in .env.secrets
+DEFAULT_LLM_PROVIDER=local
+
+# Local LLM (Ollama)
+OLLAMA_BASE_URL=http://ollama:11434
+OLLAMA_MODEL=llama3.3:latest
+
+# --- RAG ENGINE ---
+RAG_ENABLED=TRUE
+RAG_API_URL=http://rag_api:8000
+RAG_API_PORT=8000
+EMBEDDINGS_PROVIDER=ollama
+EMBEDDINGS_MODEL=nomic-embed-text

.env.librechat

@@ -0,0 +1,878 @@
+#=====================================================================#
+#                       LibreChat Configuration                       #
+#=====================================================================#
+# Please refer to the reference documentation for assistance          #
+# with configuring your LibreChat environment.                        #
+#                                                                     #
+# https://www.librechat.ai/docs/configuration/dotenv                  #
+#=====================================================================#
+
+#==================================================#
+#               Server Configuration               #
+#==================================================#
+
+HOST=localhost
+PORT=3080
+
+MONGO_URI=mongodb://127.0.0.1:27017/LibreChat
+#The maximum number of connections in the connection pool. */
+MONGO_MAX_POOL_SIZE=
+#The minimum number of connections in the connection pool. */
+MONGO_MIN_POOL_SIZE=
+#The maximum number of connections that may be in the process of being established concurrently by the connection pool. */
+MONGO_MAX_CONNECTING=
+#The maximum number of milliseconds that a connection can remain idle in the pool before being removed and closed. */
+MONGO_MAX_IDLE_TIME_MS=
+#The maximum time in milliseconds that a thread can wait for a connection to become available. */
+MONGO_WAIT_QUEUE_TIMEOUT_MS=
+# Set to false to disable automatic index creation for all models associated with this connection. */
+MONGO_AUTO_INDEX=
+# Set to `false` to disable Mongoose automatically calling `createCollection()` on every model created on this connection. */
+MONGO_AUTO_CREATE=
+
+DOMAIN_CLIENT=http://localhost:3080
+DOMAIN_SERVER=http://localhost:3080
+
+NO_INDEX=true
+# Use the address that is at most n number of hops away from the Express application.
+# req.socket.remoteAddress is the first hop, and the rest are looked for in the X-Forwarded-For header from right to left.
+# A value of 0 means that the first untrusted address would be req.socket.remoteAddress, i.e. there is no reverse proxy.
+# Defaulted to 1.
+TRUST_PROXY=1
+
+# Minimum password length for user authentication
+# Default: 8
+# Note: When using LDAP authentication, you may want to set this to 1 
+# to bypass local password validation, as LDAP servers handle their own
+# password policies.
+# MIN_PASSWORD_LENGTH=8
+
+# When enabled, the app will continue running after encountering uncaught exceptions
+# instead of exiting the process. Not recommended for production unless necessary.
+# CONTINUE_ON_UNCAUGHT_EXCEPTION=false
+
+#===============#
+# JSON Logging  #
+#===============#
+
+# Use when process console logs in cloud deployment like GCP/AWS
+CONSOLE_JSON=false
+
+#===============#
+# Debug Logging #
+#===============#
+
+DEBUG_LOGGING=true
+DEBUG_CONSOLE=false
+# Set to true to enable agent debug logging
+AGENT_DEBUG_LOGGING=false
+
+# Enable memory diagnostics (logs heap/RSS snapshots every 60s, auto-enabled with --inspect)
+# MEM_DIAG=true
+
+#=============#
+# Permissions #
+#=============#
+
+# UID=1000
+# GID=1000
+
+#==============#
+# Node Options #
+#==============#
+
+# NOTE: NODE_MAX_OLD_SPACE_SIZE is NOT recognized by Node.js directly.
+# This variable is used as a build argument for Docker or CI/CD workflows,
+# and is NOT used by Node.js to set the heap size at runtime.
+# To configure Node.js memory, use NODE_OPTIONS, e.g.:
+# NODE_OPTIONS="--max-old-space-size=6144"
+# See: https://nodejs.org/api/cli.html#--max-old-space-sizesize-in-mib
+NODE_MAX_OLD_SPACE_SIZE=6144
+
+#===============#
+# Configuration #
+#===============#
+# Use an absolute path, a relative path, or a URL
+
+# CONFIG_PATH="/alternative/path/to/librechat.yaml"
+
+#==================#
+# Langfuse Tracing #
+#==================#
+
+# Get Langfuse API keys for your project from the project settings page: https://cloud.langfuse.com
+
+# LANGFUSE_PUBLIC_KEY=
+# LANGFUSE_SECRET_KEY=
+# LANGFUSE_BASE_URL=
+
+#===================================================#
+#                     Endpoints                     #
+#===================================================#
+
+# ENDPOINTS=openAI,assistants,azureOpenAI,google,anthropic
+
+PROXY=
+
+#===================================#
+# Known Endpoints - librechat.yaml  #
+#===================================#
+# https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints
+
+# ANYSCALE_API_KEY=
+# APIPIE_API_KEY=
+# COHERE_API_KEY=
+# DEEPSEEK_API_KEY=
+# DATABRICKS_API_KEY=
+# FIREWORKS_API_KEY=
+# GROQ_API_KEY=
+# HUGGINGFACE_TOKEN=
+# MISTRAL_API_KEY=
+# OPENROUTER_KEY=
+# PERPLEXITY_API_KEY=
+# SHUTTLEAI_API_KEY=
+# TOGETHERAI_API_KEY=
+# UNIFY_API_KEY=
+# XAI_API_KEY=
+
+#============#
+# Anthropic  #
+#============#
+
+ANTHROPIC_API_KEY=user_provided
+# ANTHROPIC_MODELS=claude-sonnet-4-6,claude-opus-4-6,claude-opus-4-20250514,claude-sonnet-4-20250514,claude-3-7-sonnet-20250219,claude-3-5-sonnet-20241022,claude-3-5-haiku-20241022,claude-3-opus-20240229,claude-3-sonnet-20240229,claude-3-haiku-20240307
+# ANTHROPIC_REVERSE_PROXY=
+
+# Set to true to use Anthropic models through Google Vertex AI instead of direct API
+# ANTHROPIC_USE_VERTEX=
+# ANTHROPIC_VERTEX_REGION=us-east5
+
+#============#
+# Azure      #
+#============#
+
+# Note: these variables are DEPRECATED
+# Use the `librechat.yaml` configuration for `azureOpenAI` instead
+# You may also continue to use them if you opt out of using the `librechat.yaml` configuration
+
+# AZURE_OPENAI_DEFAULT_MODEL=gpt-3.5-turbo # Deprecated
+# AZURE_OPENAI_MODELS=gpt-3.5-turbo,gpt-4 # Deprecated
+# AZURE_USE_MODEL_AS_DEPLOYMENT_NAME=TRUE # Deprecated
+# AZURE_API_KEY= # Deprecated
+# AZURE_OPENAI_API_INSTANCE_NAME= # Deprecated
+# AZURE_OPENAI_API_DEPLOYMENT_NAME= # Deprecated
+# AZURE_OPENAI_API_VERSION= # Deprecated
+# AZURE_OPENAI_API_COMPLETIONS_DEPLOYMENT_NAME= # Deprecated
+# AZURE_OPENAI_API_EMBEDDINGS_DEPLOYMENT_NAME= # Deprecated
+
+#=================#
+#   AWS Bedrock   #
+#=================#
+
+# BEDROCK_AWS_DEFAULT_REGION=us-east-1 # A default region must be provided
+# BEDROCK_AWS_ACCESS_KEY_ID=someAccessKey
+# BEDROCK_AWS_SECRET_ACCESS_KEY=someSecretAccessKey
+# BEDROCK_AWS_SESSION_TOKEN=someSessionToken
+
+# Note: This example list is not meant to be exhaustive. If omitted, all known, supported model IDs will be included for you.
+# BEDROCK_AWS_MODELS=anthropic.claude-sonnet-4-6,anthropic.claude-opus-4-6-v1,anthropic.claude-3-5-sonnet-20240620-v1:0,meta.llama3-1-8b-instruct-v1:0
+# Cross-region inference model IDs: us.anthropic.claude-sonnet-4-6,us.anthropic.claude-opus-4-6-v1,global.anthropic.claude-opus-4-6-v1
+
+# See all Bedrock model IDs here: https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html#model-ids-arns
+
+# Notes on specific models:
+# The following models are not support due to not supporting streaming:
+# ai21.j2-mid-v1
+
+# The following models are not support due to not supporting conversation history:
+# ai21.j2-ultra-v1, cohere.command-text-v14, cohere.command-light-text-v14
+
+#============#
+# Google     #
+#============#
+
+GOOGLE_KEY=user_provided
+
+# GOOGLE_REVERSE_PROXY=
+# Some reverse proxies do not support the X-goog-api-key header, uncomment to pass the API key in Authorization header instead.
+# GOOGLE_AUTH_HEADER=true
+
+# Gemini API (AI Studio)
+# GOOGLE_MODELS=gemini-3.1-pro-preview,gemini-3.1-pro-preview-customtools,gemini-3.1-flash-lite-preview,gemini-2.5-pro,gemini-2.5-flash,gemini-2.5-flash-lite,gemini-2.0-flash,gemini-2.0-flash-lite
+
+# Vertex AI
+# GOOGLE_MODELS=gemini-3.1-pro-preview,gemini-3.1-pro-preview-customtools,gemini-3.1-flash-lite-preview,gemini-2.5-pro,gemini-2.5-flash,gemini-2.5-flash-lite,gemini-2.0-flash-001,gemini-2.0-flash-lite-001
+
+# GOOGLE_TITLE_MODEL=gemini-2.0-flash-lite-001
+
+# Google Cloud region for Vertex AI (used by both chat and image generation)
+# GOOGLE_LOC=us-central1
+
+# Alternative region env var for Gemini Image Generation
+# GOOGLE_CLOUD_LOCATION=global
+
+# Vertex AI Service Account Configuration
+# Path to your Google Cloud service account JSON file
+# GOOGLE_SERVICE_KEY_FILE=/path/to/service-account.json
+
+# Google Safety Settings
+# NOTE: These settings apply to both Vertex AI and Gemini API (AI Studio)
+#
+# For Vertex AI:
+# To use the BLOCK_NONE setting, you need either:
+# (a) Access through an allowlist via your Google account team, or
+# (b) Switch to monthly invoiced billing: https://cloud.google.com/billing/docs/how-to/invoiced-billing
+#
+# For Gemini API (AI Studio):
+# BLOCK_NONE is available by default, no special account requirements.
+#
+# Available options: BLOCK_NONE, BLOCK_ONLY_HIGH, BLOCK_MEDIUM_AND_ABOVE, BLOCK_LOW_AND_ABOVE
+#
+# GOOGLE_SAFETY_SEXUALLY_EXPLICIT=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_HATE_SPEECH=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_HARASSMENT=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_DANGEROUS_CONTENT=BLOCK_ONLY_HIGH
+# GOOGLE_SAFETY_CIVIC_INTEGRITY=BLOCK_ONLY_HIGH
+
+#========================#
+# Gemini Image Generation #
+#========================#
+
+# Gemini Image Generation Tool (for Agents)
+# Supports multiple authentication methods in priority order:
+# 1. User-provided API key (via GUI)
+# 2. GEMINI_API_KEY env var (admin-configured)
+# 3. GOOGLE_KEY env var (shared with Google chat endpoint)
+# 4. Vertex AI service account (via GOOGLE_SERVICE_KEY_FILE)
+
+# Option A: Use dedicated Gemini API key for image generation
+# GEMINI_API_KEY=your-gemini-api-key
+
+# Vertex AI model for image generation (defaults to gemini-2.5-flash-image)
+# GEMINI_IMAGE_MODEL=gemini-2.5-flash-image
+
+#============#
+# OpenAI     #
+#============#
+
+OPENAI_API_KEY=user_provided
+# OPENAI_MODELS=gpt-5,gpt-5-codex,gpt-5-mini,gpt-5-nano,o3-pro,o3,o4-mini,gpt-4.1,gpt-4.1-mini,gpt-4.1-nano,o3-mini,o1-pro,o1,gpt-4o,gpt-4o-mini
+
+DEBUG_OPENAI=false
+
+# TITLE_CONVO=false
+# OPENAI_TITLE_MODEL=gpt-4o-mini
+
+# OPENAI_SUMMARIZE=true
+# OPENAI_SUMMARY_MODEL=gpt-4o-mini
+
+# OPENAI_FORCE_PROMPT=true
+
+# OPENAI_REVERSE_PROXY=
+
+# OPENAI_ORGANIZATION=
+
+#====================#
+#   Assistants API   #
+#====================#
+
+ASSISTANTS_API_KEY=user_provided
+# ASSISTANTS_BASE_URL=
+# ASSISTANTS_MODELS=gpt-4o,gpt-4o-mini,gpt-3.5-turbo-0125,gpt-3.5-turbo-16k-0613,gpt-3.5-turbo-16k,gpt-3.5-turbo,gpt-4,gpt-4-0314,gpt-4-32k-0314,gpt-4-0613,gpt-3.5-turbo-0613,gpt-3.5-turbo-1106,gpt-4-0125-preview,gpt-4-turbo-preview,gpt-4-1106-preview
+
+#==========================#
+#   Azure Assistants API   #
+#==========================#
+
+# Note: You should map your credentials with custom variables according to your Azure OpenAI Configuration
+# The models for Azure Assistants are also determined by your Azure OpenAI configuration.
+
+# More info, including how to enable use of Assistants with Azure here:
+# https://www.librechat.ai/docs/configuration/librechat_yaml/ai_endpoints/azure#using-assistants-with-azure
+
+CREDS_KEY=f34be427ebb29de8d88c107a71546019685ed8b241d8f2ed00c3df97ad2566f0
+CREDS_IV=e2341419ec3dd3d19b13a1a87fafcbfb
+
+# Azure AI Search
+#-----------------
+AZURE_AI_SEARCH_SERVICE_ENDPOINT=
+AZURE_AI_SEARCH_INDEX_NAME=
+AZURE_AI_SEARCH_API_KEY=
+
+AZURE_AI_SEARCH_API_VERSION=
+AZURE_AI_SEARCH_SEARCH_OPTION_QUERY_TYPE=
+AZURE_AI_SEARCH_SEARCH_OPTION_TOP=
+AZURE_AI_SEARCH_SEARCH_OPTION_SELECT=
+
+# OpenAI Image Tools Customization
+#----------------
+# IMAGE_GEN_OAI_API_KEY= # Create or reuse OpenAI API key for image generation tool
+# IMAGE_GEN_OAI_BASEURL= # Custom OpenAI base URL for image generation tool
+# IMAGE_GEN_OAI_AZURE_API_VERSION= # Custom Azure OpenAI deployments
+# IMAGE_GEN_OAI_MODEL=gpt-image-1 # OpenAI image model (e.g., gpt-image-1, gpt-image-1.5)
+# IMAGE_GEN_OAI_DESCRIPTION=
+# IMAGE_GEN_OAI_DESCRIPTION_WITH_FILES=Custom description for image generation tool when files are present
+# IMAGE_GEN_OAI_DESCRIPTION_NO_FILES=Custom description for image generation tool when no files are present
+# IMAGE_EDIT_OAI_DESCRIPTION=Custom description for image editing tool
+# IMAGE_GEN_OAI_PROMPT_DESCRIPTION=Custom prompt description for image generation tool
+# IMAGE_EDIT_OAI_PROMPT_DESCRIPTION=Custom prompt description for image editing tool
+
+# DALL·E
+#----------------
+# DALLE_API_KEY=
+# DALLE3_API_KEY=
+# DALLE2_API_KEY=
+# DALLE3_SYSTEM_PROMPT=
+# DALLE2_SYSTEM_PROMPT=
+# DALLE_REVERSE_PROXY=
+# DALLE3_BASEURL=
+# DALLE2_BASEURL=
+
+# DALL·E (via Azure OpenAI)
+# Note: requires some of the variables above to be set
+#----------------
+# DALLE3_AZURE_API_VERSION=
+# DALLE2_AZURE_API_VERSION=
+
+# Flux
+#-----------------
+FLUX_API_BASE_URL=https://api.us1.bfl.ai
+# FLUX_API_BASE_URL = 'https://api.bfl.ml';
+
+# Get your API key at https://api.us1.bfl.ai/auth/profile
+# FLUX_API_KEY=
+
+# Google
+#-----------------
+GOOGLE_SEARCH_API_KEY=
+GOOGLE_CSE_ID=
+
+# Stable Diffusion
+#-----------------
+SD_WEBUI_URL=http://host.docker.internal:7860
+
+# Tavily
+#-----------------
+TAVILY_API_KEY=
+
+# Traversaal
+#-----------------
+TRAVERSAAL_API_KEY=
+
+# WolframAlpha
+#-----------------
+WOLFRAM_APP_ID=
+
+# Zapier
+#-----------------
+ZAPIER_NLA_API_KEY=
+
+#==================================================#
+#                      Search                      #
+#==================================================#
+
+SEARCH=true
+MEILI_NO_ANALYTICS=true
+MEILI_HOST=http://0.0.0.0:7700
+MEILI_MASTER_KEY=DrhYf7zENyR6AlUCKmnz0eYASOQdl6zxH7s7MKFSfFCt
+
+# Optional: Disable indexing, useful in a multi-node setup
+# where only one instance should perform an index sync.
+# MEILI_NO_SYNC=true
+
+#==================================================#
+#          Speech to Text & Text to Speech         #
+#==================================================#
+
+STT_API_KEY=
+TTS_API_KEY=
+
+#==================================================#
+#                        RAG                       #
+#==================================================#
+# More info: https://www.librechat.ai/docs/configuration/rag_api
+
+# RAG_OPENAI_BASEURL=
+# RAG_OPENAI_API_KEY=
+# RAG_USE_FULL_CONTEXT=
+# EMBEDDINGS_PROVIDER=openai
+# EMBEDDINGS_MODEL=text-embedding-3-small
+
+#===================================================#
+#                    User System                    #
+#===================================================#
+
+#========================#
+# Moderation             #
+#========================#
+
+OPENAI_MODERATION=false
+OPENAI_MODERATION_API_KEY=
+# OPENAI_MODERATION_REVERSE_PROXY=
+
+BAN_VIOLATIONS=true
+BAN_DURATION=1000 * 60 * 60 * 2
+BAN_INTERVAL=20
+
+LOGIN_VIOLATION_SCORE=1
+REGISTRATION_VIOLATION_SCORE=1
+CONCURRENT_VIOLATION_SCORE=1
+MESSAGE_VIOLATION_SCORE=1
+NON_BROWSER_VIOLATION_SCORE=20
+TTS_VIOLATION_SCORE=0
+STT_VIOLATION_SCORE=0
+FORK_VIOLATION_SCORE=0
+IMPORT_VIOLATION_SCORE=0
+FILE_UPLOAD_VIOLATION_SCORE=0
+
+LOGIN_MAX=7
+LOGIN_WINDOW=5
+REGISTER_MAX=5
+REGISTER_WINDOW=60
+
+LIMIT_CONCURRENT_MESSAGES=true
+CONCURRENT_MESSAGE_MAX=2
+
+LIMIT_MESSAGE_IP=true
+MESSAGE_IP_MAX=40
+MESSAGE_IP_WINDOW=1
+
+LIMIT_MESSAGE_USER=false
+MESSAGE_USER_MAX=40
+MESSAGE_USER_WINDOW=1
+
+ILLEGAL_MODEL_REQ_SCORE=5
+
+#========================#
+# Balance                #
+#========================#
+
+# CHECK_BALANCE=false
+# START_BALANCE=20000 # note: the number of tokens that will be credited after registration.
+
+#========================#
+# Registration and Login #
+#========================#
+
+ALLOW_EMAIL_LOGIN=true
+ALLOW_REGISTRATION=true
+ALLOW_SOCIAL_LOGIN=false
+ALLOW_SOCIAL_REGISTRATION=false
+ALLOW_PASSWORD_RESET=false
+# ALLOW_ACCOUNT_DELETION=true # note: enabled by default if omitted/commented out
+ALLOW_UNVERIFIED_EMAIL_LOGIN=true
+
+SESSION_EXPIRY=1000 * 60 * 15
+REFRESH_TOKEN_EXPIRY=(1000 * 60 * 60 * 24) * 7
+
+JWT_SECRET=16f8c0ef4a5d391b26034086c628469d3f9f497f08163ab9b40137092f2909ef
+JWT_REFRESH_SECRET=eaa5191f2914e30b9387fd84e254e4ba6fc51b4654968a9b0803b456a54b8418
+
+# Discord
+DISCORD_CLIENT_ID=
+DISCORD_CLIENT_SECRET=
+DISCORD_CALLBACK_URL=/oauth/discord/callback
+
+# Facebook
+FACEBOOK_CLIENT_ID=
+FACEBOOK_CLIENT_SECRET=
+FACEBOOK_CALLBACK_URL=/oauth/facebook/callback
+
+# GitHub
+GITHUB_CLIENT_ID=
+GITHUB_CLIENT_SECRET=
+GITHUB_CALLBACK_URL=/oauth/github/callback
+# GitHub Enterprise
+# GITHUB_ENTERPRISE_BASE_URL=
+# GITHUB_ENTERPRISE_USER_AGENT=
+
+# Google
+GOOGLE_CLIENT_ID=
+GOOGLE_CLIENT_SECRET=
+GOOGLE_CALLBACK_URL=/oauth/google/callback
+
+# Apple
+APPLE_CLIENT_ID=
+APPLE_TEAM_ID=
+APPLE_KEY_ID=
+APPLE_PRIVATE_KEY_PATH=
+APPLE_CALLBACK_URL=/oauth/apple/callback
+
+# OpenID
+OPENID_CLIENT_ID=
+OPENID_CLIENT_SECRET=
+OPENID_ISSUER=
+OPENID_SESSION_SECRET=
+OPENID_SCOPE="openid profile email"
+OPENID_CALLBACK_URL=/oauth/openid/callback
+OPENID_REQUIRED_ROLE=
+OPENID_REQUIRED_ROLE_TOKEN_KIND=
+OPENID_REQUIRED_ROLE_PARAMETER_PATH=
+OPENID_ADMIN_ROLE=
+OPENID_ADMIN_ROLE_PARAMETER_PATH=
+OPENID_ADMIN_ROLE_TOKEN_KIND=
+# Set to determine which user info property returned from OpenID Provider to store as the User's username
+OPENID_USERNAME_CLAIM=
+# Set to determine which user info property returned from OpenID Provider to store as the User's name
+OPENID_NAME_CLAIM=
+# Set to determine which user info claim to use as the email/identifier for user matching (e.g., "upn" for Entra ID)
+# When not set, defaults to: email -> preferred_username -> upn
+OPENID_EMAIL_CLAIM=
+# Optional audience parameter for OpenID authorization requests
+OPENID_AUDIENCE=
+
+OPENID_BUTTON_LABEL=
+OPENID_IMAGE_URL=
+# Set to true to automatically redirect to the OpenID provider when a user visits the login page
+# This will bypass the login form completely for users, only use this if OpenID is your only authentication method
+OPENID_AUTO_REDIRECT=false
+# Set to true to use PKCE (Proof Key for Code Exchange) for OpenID authentication
+OPENID_USE_PKCE=false
+#Set to true to reuse openid tokens for authentication management instead of using the mongodb session and the custom refresh token.
+OPENID_REUSE_TOKENS=
+#By default, signing key verification results are cached in order to prevent excessive HTTP requests to the JWKS endpoint.
+#If a signing key matching the kid is found, this will be cached and the next time this kid is requested the signing key will be served from the cache.
+#Default is true.
+OPENID_JWKS_URL_CACHE_ENABLED=
+OPENID_JWKS_URL_CACHE_TIME= # 600000 ms eq to 10 minutes leave empty to disable caching
+#Set to true to trigger token exchange flow to acquire access token for the userinfo endpoint.
+OPENID_ON_BEHALF_FLOW_FOR_USERINFO_REQUIRED=
+OPENID_ON_BEHALF_FLOW_USERINFO_SCOPE="user.read" # example for Scope Needed for Microsoft Graph API
+# Set to true to use the OpenID Connect end session endpoint for logout
+OPENID_USE_END_SESSION_ENDPOINT=
+# URL to redirect to after OpenID logout (defaults to ${DOMAIN_CLIENT}/login)
+OPENID_POST_LOGOUT_REDIRECT_URI=
+# Maximum logout URL length before using logout_hint instead of id_token_hint (default: 2000)
+OPENID_MAX_LOGOUT_URL_LENGTH=
+
+#========================#
+# SharePoint Integration #
+#========================#
+# Requires Entra ID (OpenID) authentication to be configured
+
+# Enable SharePoint file picker in chat and agent panels
+# ENABLE_SHAREPOINT_FILEPICKER=true
+
+# SharePoint tenant base URL (e.g., https://yourtenant.sharepoint.com)
+# SHAREPOINT_BASE_URL=https://yourtenant.sharepoint.com
+
+# Microsoft Graph API And SharePoint scopes for file picker
+# SHAREPOINT_PICKER_SHAREPOINT_SCOPE==https://yourtenant.sharepoint.com/AllSites.Read
+# SHAREPOINT_PICKER_GRAPH_SCOPE=Files.Read.All
+#========================#
+
+# SAML
+# Note: If OpenID is enabled, SAML authentication will be automatically disabled.
+SAML_ENTRY_POINT=
+SAML_ISSUER=
+SAML_CERT=
+SAML_CALLBACK_URL=/oauth/saml/callback
+SAML_SESSION_SECRET=
+
+# Attribute mappings (optional)
+SAML_EMAIL_CLAIM=
+SAML_USERNAME_CLAIM=
+SAML_GIVEN_NAME_CLAIM=
+SAML_FAMILY_NAME_CLAIM=
+SAML_PICTURE_CLAIM=
+SAML_NAME_CLAIM=
+
+# Logint buttion settings (optional)
+SAML_BUTTON_LABEL=
+SAML_IMAGE_URL=
+
+# Whether the SAML Response should be signed.
+# - If "true", the entire `SAML Response` will be signed.
+# - If "false" or unset, only the `SAML Assertion` will be signed (default behavior).
+# SAML_USE_AUTHN_RESPONSE_SIGNED=
+
+
+#===============================================#
+# Microsoft Graph API / Entra ID Integration  #
+#===============================================#
+
+# Enable Entra ID people search integration in permissions/sharing system
+# When enabled, the people picker will search both local database and Entra ID
+USE_ENTRA_ID_FOR_PEOPLE_SEARCH=false
+
+# When enabled, entra id groups owners will be considered as members of the group
+ENTRA_ID_INCLUDE_OWNERS_AS_MEMBERS=false
+
+# Microsoft Graph API scopes needed for people/group search
+# Default scopes provide access to user profiles and group memberships
+OPENID_GRAPH_SCOPES=User.Read,People.Read,GroupMember.Read.All
+
+# LDAP
+LDAP_URL=
+LDAP_BIND_DN=
+LDAP_BIND_CREDENTIALS=
+LDAP_USER_SEARCH_BASE=
+#LDAP_SEARCH_FILTER="mail="
+LDAP_CA_CERT_PATH=
+# LDAP_TLS_REJECT_UNAUTHORIZED=
+# LDAP_STARTTLS=
+# LDAP_LOGIN_USES_USERNAME=true
+# LDAP_ID=
+# LDAP_USERNAME=
+# LDAP_EMAIL=
+# LDAP_FULL_NAME=
+
+#========================#
+# Email Password Reset   #
+#========================#
+
+EMAIL_SERVICE=
+EMAIL_HOST=
+EMAIL_PORT=25
+EMAIL_ENCRYPTION=
+EMAIL_ENCRYPTION_HOSTNAME=
+EMAIL_ALLOW_SELFSIGNED=
+# Leave both empty for SMTP servers that do not require authentication
+EMAIL_USERNAME=
+EMAIL_PASSWORD=
+EMAIL_FROM_NAME=
+EMAIL_FROM=noreply@librechat.ai
+
+#========================#
+#      Mailgun API       #
+#========================#
+
+# MAILGUN_API_KEY=your-mailgun-api-key
+# MAILGUN_DOMAIN=mg.yourdomain.com
+# EMAIL_FROM=noreply@yourdomain.com
+# EMAIL_FROM_NAME="LibreChat"
+
+# # Optional: For EU region
+# MAILGUN_HOST=https://api.eu.mailgun.net
+
+#========================#
+# Firebase CDN           #
+#========================#
+
+FIREBASE_API_KEY=
+FIREBASE_AUTH_DOMAIN=
+FIREBASE_PROJECT_ID=
+FIREBASE_STORAGE_BUCKET=
+FIREBASE_MESSAGING_SENDER_ID=
+FIREBASE_APP_ID=
+
+#========================#
+# S3 AWS Bucket          #
+#========================#
+
+AWS_ENDPOINT_URL=
+AWS_ACCESS_KEY_ID=
+AWS_SECRET_ACCESS_KEY=
+AWS_REGION=
+AWS_BUCKET_NAME=
+# Required for path-style S3-compatible providers (MinIO, Hetzner, Backblaze B2, etc.)
+# that don't support virtual-hosted-style URLs (bucket.endpoint). Not needed for AWS S3.
+# AWS_FORCE_PATH_STYLE=false
+
+#========================#
+# Azure Blob Storage     #
+#========================#
+
+AZURE_STORAGE_CONNECTION_STRING=
+AZURE_STORAGE_PUBLIC_ACCESS=false
+AZURE_CONTAINER_NAME=files
+
+#========================#
+# Shared Links           #
+#========================#
+
+ALLOW_SHARED_LINKS=true
+# Allows unauthenticated access to shared links. Defaults to false (auth required) if not set.
+ALLOW_SHARED_LINKS_PUBLIC=false
+
+#==============================#
+# Static File Cache Control    #
+#==============================#
+
+# Leave commented out to use defaults: 1 day (86400 seconds) for s-maxage and 2 days (172800 seconds) for max-age
+# NODE_ENV must be set to production for these to take effect
+# STATIC_CACHE_MAX_AGE=172800
+# STATIC_CACHE_S_MAX_AGE=86400
+
+# If you have another service in front of your LibreChat doing compression, disable express based compression here
+# DISABLE_COMPRESSION=true
+
+# If you have gzipped version of uploaded image images in the same folder, this will enable gzip scan and serving of these images
+# Note: The images folder will be scanned on startup and a ma kept in memory. Be careful for large number of images.
+# ENABLE_IMAGE_OUTPUT_GZIP_SCAN=true
+
+#===================================================#
+#                        UI                         #
+#===================================================#
+
+APP_TITLE=LibreChat
+# CUSTOM_FOOTER="My custom footer"
+HELP_AND_FAQ_URL=https://librechat.ai
+
+# SHOW_BIRTHDAY_ICON=true
+
+# Google tag manager id
+#ANALYTICS_GTM_ID=user provided google tag manager id
+
+# limit conversation file imports to a certain number of bytes in size to avoid the container
+# maxing out memory limitations by unremarking this line and supplying a file size in bytes
+# such as the below example of 250 mib
+# CONVERSATION_IMPORT_MAX_FILE_SIZE_BYTES=262144000
+
+
+#===============#
+# REDIS Options #
+#===============#
+
+# Enable Redis for caching and session storage
+# USE_REDIS=true
+# Enable Redis for resumable LLM streams (defaults to USE_REDIS value if not set)
+# Set to false to use in-memory storage for streams while keeping Redis for other caches
+# USE_REDIS_STREAMS=true
+
+# Single Redis instance
+# REDIS_URI=redis://127.0.0.1:6379
+
+# Redis cluster (multiple nodes)
+# REDIS_URI=redis://127.0.0.1:7001,redis://127.0.0.1:7002,redis://127.0.0.1:7003
+
+# Redis with TLS/SSL encryption and CA certificate
+# REDIS_URI=rediss://127.0.0.1:6380
+# REDIS_CA=/path/to/ca-cert.pem
+
+# Elasticache may need to use an alternate dnsLookup for TLS connections.  see "Special Note: Aws Elasticache Clusters with TLS" on this webpage: https://www.npmjs.com/package/ioredis
+# Enable alternative dnsLookup for redis
+# REDIS_USE_ALTERNATIVE_DNS_LOOKUP=true
+
+# Redis authentication (if required)
+# REDIS_USERNAME=your_redis_username
+# REDIS_PASSWORD=your_redis_password
+
+# Redis key prefix configuration
+# Use environment variable name for dynamic prefix (recommended for cloud deployments)
+# REDIS_KEY_PREFIX_VAR=K_REVISION
+# Or use static prefix directly
+# REDIS_KEY_PREFIX=librechat
+
+# Redis connection limits
+# REDIS_MAX_LISTENERS=40
+
+# Redis ping interval in seconds (0 = disabled, >0 = enabled)
+# When set to a positive integer, Redis clients will ping the server at this interval to keep connections alive
+# When unset or 0, no pinging is performed (recommended for most use cases)
+# REDIS_PING_INTERVAL=300
+
+# Force specific cache namespaces to use in-memory storage even when Redis is enabled
+# Comma-separated list of CacheKeys
+# Defaults to CONFIG_STORE,APP_CONFIG so YAML-derived config stays per-container (safe for blue/green deployments)
+# Set to empty string to force all namespaces through Redis: FORCED_IN_MEMORY_CACHE_NAMESPACES=
+# FORCED_IN_MEMORY_CACHE_NAMESPACES=CONFIG_STORE,APP_CONFIG
+
+# Leader Election Configuration (for multi-instance deployments with Redis)
+# Duration in seconds that the leader lease is valid before it expires (default: 25)
+# LEADER_LEASE_DURATION=25
+# Interval in seconds at which the leader renews its lease (default: 10)
+# LEADER_RENEW_INTERVAL=10
+# Maximum number of retry attempts when renewing the lease fails (default: 3)
+# LEADER_RENEW_ATTEMPTS=3
+# Delay in seconds between retry attempts when renewing the lease (default: 0.5)
+# LEADER_RENEW_RETRY_DELAY=0.5
+
+#==================================================#
+#                      Others                      #
+#==================================================#
+#   You should leave the following commented out   #
+
+# NODE_ENV=
+
+# E2E_USER_EMAIL=
+# E2E_USER_PASSWORD=
+
+#=====================================================#
+#                  Cache Headers                      #
+#=====================================================#
+#   Headers that control caching of the index.html    #
+#   Default configuration prevents caching to ensure  #
+#   users always get the latest version. Customize    #
+#   only if you understand caching implications.      #
+
+# INDEX_CACHE_CONTROL=no-cache, no-store, must-revalidate
+# INDEX_PRAGMA=no-cache
+# INDEX_EXPIRES=0
+
+# no-cache: Forces validation with server before using cached version
+# no-store: Prevents storing the response entirely
+# must-revalidate: Prevents using stale content when offline
+
+#=====================================================#
+#                  OpenWeather                        #
+#=====================================================#
+OPENWEATHER_API_KEY=
+
+#====================================#
+# LibreChat Code Interpreter API     #
+#====================================#
+
+# https://code.librechat.ai
+# LIBRECHAT_CODE_API_KEY=your-key
+
+#======================#
+# Web Search           #
+#======================#
+
+# Note: All of the following variable names can be customized.
+# Omit values to allow user to provide them.
+
+# For more information on configuration values, see:
+# https://librechat.ai/docs/features/web_search
+
+# Search Provider (Required)
+# SERPER_API_KEY=your_serper_api_key
+
+# Scraper (Required)
+# FIRECRAWL_API_KEY=your_firecrawl_api_key
+# Optional: Custom Firecrawl API URL
+# FIRECRAWL_API_URL=your_firecrawl_api_url
+
+# Reranker (Required)
+# JINA_API_KEY=your_jina_api_key
+# or
+# COHERE_API_KEY=your_cohere_api_key
+
+#======================#
+# MCP Configuration    #
+#======================#
+
+# Treat 401/403 responses as OAuth requirement when no oauth metadata found
+# MCP_OAUTH_ON_AUTH_ERROR=true
+
+# Timeout for OAuth detection requests in milliseconds
+# MCP_OAUTH_DETECTION_TIMEOUT=5000
+
+# Cache connection status checks for this many milliseconds to avoid expensive verification
+# MCP_CONNECTION_CHECK_TTL=60000
+
+# Skip code challenge method validation (e.g., for AWS Cognito that supports S256 but doesn't advertise it)
+# When set to true, forces S256 code challenge even if not advertised in .well-known/openid-configuration
+# MCP_SKIP_CODE_CHALLENGE_CHECK=false
+
+# Circuit breaker: max connect/disconnect cycles before tripping (per server)
+# MCP_CB_MAX_CYCLES=7
+
+# Circuit breaker: sliding window (ms) for counting cycles
+# MCP_CB_CYCLE_WINDOW_MS=45000
+
+# Circuit breaker: cooldown (ms) after the cycle breaker trips
+# MCP_CB_CYCLE_COOLDOWN_MS=15000
+
+# Circuit breaker: max consecutive failed connection rounds before backoff
+# MCP_CB_MAX_FAILED_ROUNDS=3
+
+# Circuit breaker: sliding window (ms) for counting failed rounds
+# MCP_CB_FAILED_WINDOW_MS=120000
+
+# Circuit breaker: base backoff (ms) after failed round threshold is reached
+# MCP_CB_BASE_BACKOFF_MS=30000
+
+# Circuit breaker: max backoff cap (ms) for exponential backoff
+# MCP_CB_MAX_BACKOFF_MS=300000

.env.make

@@ -0,0 +1,8 @@
+# Auto-generated from pyproject.toml — do not edit manually
+ALFRED_VERSION=0.1.7
+PYTHON_VERSION=3.14.3
+IMAGE_NAME=alfred_media_organizer
+SERVICE_NAME=alfred
+LIBRECHAT_VERSION=v0.8.4
+RAG_VERSION=v0.7.3
+UV_VERSION=0.11.6

.gitea/workflows/ci.yml

@@ -2,11 +2,10 @@ name: CI/CD Awesome Pipeline
 
 on:
   push:
-    branches: [main]
     tags:
       - 'v*.*.*'
-  pull_request:
+
-    branches: [main]
+  workflow_dispatch:
 
 env:
   REGISTRY_URL: ${{ vars.REGISTRY_URL || 'gitea.iswearihadsomethingforthis.net' }}
@@ -30,34 +29,64 @@ jobs:
     name: Build & Push to Registry
     runs-on: ubuntu-latest
     needs: test
-    if: startsWith(github.ref, 'refs/tags/')
-    steps:
-      - name: Debug ref
-        run: |
-          echo "github.ref = ${{ github.ref }}"
-          echo "GITHUB_REF = $GITHUB_REF"
-          echo "This should only run on tags!"
 
+    steps:
       - name: Checkout code
         uses: actions/checkout@v4
 
+      - name: Generate build variables
+        run: python scripts/generate_build_vars.py
+
       - name: Load config from Makefile
         id: config
-        run: |
+        run: make -s _ci-dump-config >> $GITHUB_OUTPUT
-          eval "$(make _ci-image-name)"
-          echo "image_name=${IMAGE_NAME}" >> $GITHUB_OUTPUT
 
-      - name: Extract version from tag
+      - name: 🏷️ Docker Metadata (Tags & Labels)
-        id: version
+        id: meta
-        run: echo "version=${GITHUB_REF#refs/tags/v}" >> $GITHUB_OUTPUT
+        uses: docker/metadata-action@v5
+        with:
+          images: gitea.iswearihadsomethingforthis.net/francwa/${{ steps.config.outputs.image_name }}
+          tags: |
+            # Tagged (v1.2.3)
+            type=semver,pattern={{ version }}
+            # Latest (main)
+            type=raw,value=latest,enable={{ is_default_branch }}
+            # Feature branches
+            type=ref,event=branch
 
-      - name: Build production image
+      - name: Login to Gitea Registry
-        run: make build
+        uses: docker/login-action@v3
+        with:
+          registry: gitea.iswearihadsomethingforthis.net
+          username: ${{ gitea.actor }}
+          password: ${{ secrets.G1T34_TOKEN }}
 
-      - name: Tag and push to registry
+      - name: Build and push
-        run: |
+        id: docker_build
-          docker tag ${{ steps.config.outputs.image_name }}:latest ${{ env.REGISTRY_URL }}/${{ env.REGISTRY_USER }}/${{ steps.config.outputs.image_name }}:${{ steps.version.outputs.version }}
+        uses: docker/build-push-action@v5
-          docker tag ${{ steps.config.outputs.image_name }}:latest ${{ env.REGISTRY_URL }}/${{ env.REGISTRY_USER }}/${{ steps.config.outputs.image_name }}:latest
+        with:
-          echo "${{ secrets.G1T34_TOKEN }}" | docker login ${{ env.REGISTRY_URL }} -u ${{ env.REGISTRY_USER }} --password-stdin
+          context: .
-          docker push ${{ env.REGISTRY_URL }}/${{ env.REGISTRY_USER }}/${{ steps.config.outputs.image_name }}:${{ steps.version.outputs.version }}
+          push: true
-          docker push ${{ env.REGISTRY_URL }}/${{ env.REGISTRY_USER }}/${{ steps.config.outputs.image_name }}:latest
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          build-args: |
+            PYTHON_VERSION=${{ steps.config.outputs.python_version }}
+            PYTHON_VERSION_SHORT=${{ steps.config.outputs.python_version_short }}
+            RUNNER=${{ steps.config.outputs.runner }}
+
+      - name: 🛡️ Run Trivy Vulnerability Scanner
+        uses: docker://aquasec/trivy:latest
+        env:
+          TRIVY_USERNAME: ${{ gitea.actor }}
+          TRIVY_PASSWORD: ${{ secrets.G1T34_TOKEN }}
+          # Unset the fake GITHUB_TOKEN injected by Gitea
+          GITHUB_TOKEN: ""
+        with:
+          args: image --format table --output trivy-report.txt --exit-code 0 --ignore-unfixed --severity CRITICAL,HIGH gitea.iswearihadsomethingforthis.net/francwa/${{ steps.config.outputs.image_name }}:latest
+
+      - name: 📤 Upload Security Report
+        uses: actions/upload-artifact@v3
+        with:
+          name: security-report
+          path: trivy-report.txt
+          retention-days: 7

.gitea/workflows/renovate.yaml

@@ -0,0 +1,22 @@
+name: Renovate Bot
+
+on:
+  schedule:
+    # Every Monday 4AM
+    - cron: '0 4 * * 1'
+  workflow_dispatch:
+
+jobs:
+  renovate:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Run Renovate
+        uses: docker://renovate/renovate:latest
+        env:
+          RENOVATE_PLATFORM: "gitea"
+          RENOVATE_ENDPOINT: "https://gitea.iswearihadsomethingforthis.net/api/v1"
+          RENOVATE_TOKEN: "${{ secrets.RENOVATE_TOKEN }}"
+          RENOVATE_REPOSITORIES: '["${{ gitea.repository }}"]'
+          RENOVATE_GIT_AUTHOR: "Renovate Bot <renovate@bot.local>"
+          # Might need a free github token if lots of depencies
+          # RENOVATE_GITHUB_TOKEN: "${{ secrets.GITHUB_COM_TOKEN }}"

.gitignore

@@ -55,7 +55,21 @@ coverage.xml
 Thumbs.db
 
 # Secrets
-.env
+.env.secrets
 
 # Backup files
 *.backup
+
+# Application data dir
+data/*
+
+# Application logs
+logs/*
+
+# Documentation folder
+docs/
+
+# .md files
+*.md
+
+#

Dockerfile

@@ -0,0 +1,91 @@
+# syntax=docker/dockerfile:1
+# check=skip=InvalidDefaultArgInFrom
+
+ARG PYTHON_VERSION
+ARG UV_VERSION
+
+# Stage 0: uv binary (workaround — --from doesn't support ARG expansion)
+FROM ghcr.io/astral-sh/uv:${UV_VERSION} AS uv-bin
+
+# ===========================================
+# Stage 1: Builder
+# ===========================================
+FROM python:${PYTHON_VERSION}-slim-bookworm AS builder
+
+ENV DEBIAN_FRONTEND=noninteractive \
+    PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    UV_PROJECT_ENVIRONMENT=/venv
+
+# Install build dependencies
+RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
+    --mount=type=cache,target=/var/lib/apt,sharing=locked \
+    apt-get update \
+    && apt-get install -y --no-install-recommends build-essential
+
+# Install uv globally
+COPY --from=uv-bin /uv /usr/local/bin/uv
+
+WORKDIR /tmp
+
+COPY pyproject.toml uv.lock Makefile ./
+
+# Install dependencies into /venv
+RUN --mount=type=cache,target=/root/.cache/uv uv sync
+
+COPY scripts/ ./scripts/
+COPY .env.example ./
+
+# ===========================================
+# Stage 2: Testing
+# ===========================================
+FROM builder AS test
+
+RUN --mount=type=cache,target=/root/.cache/uv uv sync --group dev
+
+COPY alfred/ ./alfred
+COPY scripts ./scripts
+COPY tests/  ./tests
+
+# ===========================================
+# Stage 3: Runtime
+# ===========================================
+FROM python:${PYTHON_VERSION}-slim-bookworm AS runtime
+
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    PYTHONPATH=/home/appuser \
+    PATH="/venv/bin:$PATH"
+
+# Install runtime dependencies
+RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
+    --mount=type=cache,target=/var/lib/apt,sharing=locked \
+    apt-get update \
+    && apt-get install -y --no-install-recommends ca-certificates
+
+# Create non-root user
+RUN useradd -m -u 1000 -s /bin/bash appuser
+
+# Create data directories
+RUN mkdir -p /data /logs \
+    && chown -R appuser:appuser /data /logs
+
+USER appuser
+WORKDIR /home/appuser
+
+# Copy venv from builder stage
+COPY --from=builder /venv /venv
+
+# Copy application code
+COPY --chown=appuser:appuser alfred/ ./alfred
+COPY --chown=appuser:appuser scripts/ ./scripts
+COPY --chown=appuser:appuser .env.example ./
+COPY --chown=appuser:appuser pyproject.toml ./
+
+VOLUME ["/data", "/logs"]
+EXPOSE 8000
+
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+    CMD python -c "import requests; requests.get('http://localhost:8000/health', timeout=5).raise_for_status()" || exit 1
+
+CMD ["python", "-m", "uvicorn", "alfred.app:app", "--host", "0.0.0.0", "--port", "8000"]

Makefile

@@ -1,257 +1,187 @@
-.POSIX:
-.SUFFIXES:
 .DEFAULT_GOAL := help
 
-# --- SETTINGS ---
+# --- Load Config from pyproject.toml ---
-PYTHON_VERSION = 3.12.7
+export
-PYTHON_VERSION_SHORT = $(shell echo $(PYTHON_VERSION) | cut -d. -f1,2)
+-include .env.make
-# Change to 'uv' when ready.
-RUNNER ?= poetry
 
-export PYTHON_VERSION
+# --- Profiles management ---
-export PYTHON_VERSION_SHORT
+# Usage: make up p=rag,meili
-export RUNNER
+p ?= full
-export IMAGE_NAME
+PROFILES_PARAM	:= COMPOSE_PROFILES=$(p)
 
-# --- VARIABLES ---
+# --- Commands ---
-CORE_DIR     = brain
+DOCKER_COMPOSE	:= docker compose \
-SERVICE_NAME = agent_media
+					--env-file .env.alfred \
-IMAGE_NAME   = agent_media
+					--env-file .env.secrets \
+					--env-file .env.make
+DOCKER_BUILD	:= DOCKER_BUILDKIT=1 docker build \
+					--build-arg PYTHON_VERSION=$(PYTHON_VERSION) \
+					--build-arg UV_VERSION=$(UV_VERSION)
 
-# --- ADAPTERS ---
+# --- Phony ---
-# UV uses "sync", Poetry uses "install". Both install DEV deps by default.
+.PHONY: bootstrap up down restart logs ps shell build build-test install \
-INSTALL_CMD = $(if $(filter uv,$(RUNNER)),sync,install)
+	update install-hooks test coverage lint format clean major minor patch help
 
-# --- MACROS ---
+# --- Setup ---
-ARGS           = $(filter-out $@,$(MAKECMDGOALS))
+.env.alfred .env.librechat .env.secrets .env.make:
-BUMP_CMD       = cd $(CORE_DIR) && $(RUNNER) run bump-my-version bump
+	@echo "Initializing environment..."
-COMPOSE_CMD    = docker-compose
+	@uv run python scripts/bootstrap.py \
-DOCKER_CMD     = docker build \
+		&& echo "✓ Environment ready" \
- --build-arg PYTHON_VERSION=$(PYTHON_VERSION) \
+		|| (echo "✗ Environment setup failed" && exit 1)
- --build-arg PYTHON_VERSION_SHORT=$(PYTHON_VERSION_SHORT) \
- --build-arg RUNNER=$(RUNNER) \
- -f $(CORE_DIR)/Dockerfile \
- -t $(IMAGE_NAME):latest .
 
-RUNNER_ADD     = cd $(CORE_DIR) && $(RUNNER) add
+bootstrap: .env.alfred .env.librechat .env.secrets .env.make
-RUNNER_HOOKS   = cd $(CORE_DIR) && $(RUNNER) run pre-commit install -c ../.pre-commit-config.yaml
-RUNNER_INSTALL = cd $(CORE_DIR) && $(RUNNER) $(INSTALL_CMD)
-RUNNER_RUN     = cd $(CORE_DIR) && $(RUNNER) run
-RUNNER_UPDATE  = cd $(CORE_DIR) && $(RUNNER) update
 
-# --- STYLES ---
+# --- Docker ---
-B = \033[1m
+up: .env.alfred .env.secrets
-G = \033[32m
+	@echo "Starting containers with profiles: [full]..."
-T = \033[36m
+	@$(PROFILES_PARAM) $(DOCKER_COMPOSE) up -d --remove-orphans \
-R = \033[0m
+		&& echo "✓ Containers started" \
+		|| (echo "✗ Failed to start containers" && exit 1)
 
-# --- TARGETS ---
+down:
-.PHONY: add build build-test check-docker check-runner clean coverage down format help init-dotenv install install-hooks lint logs major minor patch prune ps restart run shell test up update _check_branch _ci-image-name _ci-run-tests _push_tag
+	@echo "Stopping containers..."
+	@$(PROFILES_PARAM) $(DOCKER_COMPOSE) down \
+		&& echo "✓ Containers stopped" \
+		|| (echo "✗ Failed to stop containers" && exit 1)
 
-# Catch-all for args
+restart:
-%:
+	@echo "Restarting containers..."
-	@:
+	@$(PROFILES_PARAM) $(DOCKER_COMPOSE) restart \
+		&& echo "✓ Containers restarted" \
+		|| (echo "✗ Failed to restart containers" && exit 1)
 
-add: check-runner
+logs:
-	@echo "$(T)➕ Adding dependency ($(RUNNER)): $(ARGS)$(R)"
+	@echo "Following logs (Ctrl+C to exit)..."
-	$(RUNNER_ADD) $(ARGS)
+	@$(PROFILES_PARAM) $(DOCKER_COMPOSE) logs -f
 
-build: check-docker
+ps:
-	@echo "$(T)🐳 Building Docker image...$(R)"
+	@echo "Container status:"
-	$(DOCKER_CMD)
+	@$(PROFILES_PARAM) $(DOCKER_COMPOSE) ps
-	@echo "✅ Image $(IMAGE_NAME):latest ready."
 
-build-test: check-docker
+shell:
-	@echo "$(T)🐳 Building test image (with dev deps)...$(R)"
+	@echo "Opening shell in $(SERVICE_NAME)..."
-	docker build \
+	@$(DOCKER_COMPOSE) exec $(SERVICE_NAME) /bin/bash
-		--build-arg RUNNER=$(RUNNER) \
-		--build-arg PYTHON_VERSION=$(PYTHON_VERSION) \
-		--build-arg PYTHON_VERSION_SHORT=$(PYTHON_VERSION_SHORT) \
-		-f $(CORE_DIR)/Dockerfile \
-		--target test \
-		-t $(IMAGE_NAME):test .
-	@echo "✅ Test image $(IMAGE_NAME):test ready."
 
-check-docker:
+# --- Build ---
-	@command -v docker >/dev/null 2>&1 || { echo "$(R)❌ Docker not installed$(R)"; exit 1; }
+build: .env.make
-	@docker info >/dev/null 2>&1 || { echo "$(R)❌ Docker daemon not running$(R)"; exit 1; }
+	@echo "Building image $(IMAGE_NAME):latest ..."
+	@$(DOCKER_BUILD) -t $(IMAGE_NAME):latest . \
+		&& echo "✓ Build complete" \
+		|| (echo "✗ Build failed" && exit 1)
 
-check-runner:
+build-test: .env.make
-	@command -v $(RUNNER) >/dev/null 2>&1 || { echo "$(R)❌ $(RUNNER) not installed$(R)"; exit 1; }
+	@echo "Building test image $(IMAGE_NAME):test..."
+	@$(DOCKER_BUILD) --target test -t $(IMAGE_NAME):test . \
+		&& echo "✓ Test image built" \
+		|| (echo "✗ Build failed" && exit 1)
+
+# --- Dependencies ---
+install:
+	@echo "Installing dependencies with uv..."
+	@uv install \
+		&& echo "✓ Dependencies installed" \
+		|| (echo "✗ Installation failed" && exit 1)
+
+install-hooks:
+	@echo "Installing pre-commit hooks..."
+	@uv run pre-commit install \
+		&& echo "✓ Hooks installed" \
+		|| (echo "✗ Hook installation failed" && exit 1)
+
+update:
+	@echo "Updating dependencies with uv..."
+	@uv update \
+		&& echo "✓ Dependencies updated" \
+		|| (echo "✗ Update failed" && exit 1)
+
+# --- Quality ---
+test:
+	@echo "Running tests..."
+	@uv run pytest \
+		&& echo "✓ Tests passed" \
+		|| (echo "✗ Tests failed" && exit 1)
+
+coverage:
+	@echo "Running tests with coverage..."
+	@uv run pytest --cov=. --cov-report=html --cov-report=term \
+		&& echo "✓ Coverage report generated" \
+		|| (echo "✗ Coverage failed" && exit 1)
+
+lint:
+	@echo "Linting code..."
+	@uv run ruff check --fix . \
+		&& echo "✓ Linting complete" \
+		|| (echo "✗ Linting failed" && exit 1)
+
+format:
+	@echo "Formatting code..."
+	@uv run ruff format . && uv run ruff check --fix . \
+		&& echo "✓ Code formatted" \
+		|| (echo "✗ Formatting failed" && exit 1)
 
 clean:
-	@echo "$(T)🧹 Cleaning caches...$(R)"
+	@echo "Cleaning build artifacts..."
-	cd $(CORE_DIR) && rm -rf .ruff_cache __pycache__ .pytest_cache
+	@rm -rf .ruff_cache __pycache__ .pytest_cache htmlcov .coverage
-	find $(CORE_DIR) -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true
+	@find . -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true
-	find $(CORE_DIR) -type d -name ".pytest_cache" -exec rm -rf {} + 2>/dev/null || true
+	@echo "✓ Cleanup complete"
-	find $(CORE_DIR) -type f -name "*.pyc" -delete 2>/dev/null || true
-	@echo "✅ Caches cleaned."
 
-coverage: check-runner
+# --- Versioning ---
-	@echo "$(T)📊 Running tests with coverage...$(R)"
+major minor patch: _check-main
-	$(RUNNER_RUN) pytest --cov=. --cov-report=html --cov-report=term $(ARGS)
+	@echo "Bumping $@ version..."
-	@echo "✅ Report generated in htmlcov/"
+	@uv run bump-my-version bump $@ \
+		&& echo "✓ Version bumped" \
+		|| (echo "✗ Version bump failed" && exit 1)
 
-down: check-docker
+	@echo "Pushing tags..."
-	@echo "$(T)🛑 Stopping containers...$(R)"
+	@git push --tags \
-	$(COMPOSE_CMD) down
+		&& echo "✓ Tags pushed" \
-	@echo "✅ System stopped."
+		|| (echo "✗ Push failed" && exit 1)
 
-format: check-runner
+# CI/CD helpers
-	@echo "$(T)✨ Formatting with Ruff...$(R)"
+_ci-dump-config:
-	$(RUNNER_RUN) ruff format .
+	@echo "image_name=$(IMAGE_NAME)"
-	$(RUNNER_RUN) ruff check --fix .
+	@echo "python_version=$(PYTHON_VERSION)"
-	@echo "✅ Code cleaned."
+	@echo "uv_version=$(UV_VERSION)"
+	@echo "service_name=$(SERVICE_NAME)"
 
-help:
+_ci-run-tests:build-test
-	@echo "$(B)Available commands:$(R)"
+	@echo "Running tests in Docker..."
-	@echo ""
-	@echo "$(G)Setup:$(R)"
-	@echo "  $(T)check-docker 		$(R)  Verify Docker is installed and running."
-	@echo "  $(T)check-runner 		$(R)  Verify package manager ($(RUNNER))."
-	@echo "  $(T)init-dotenv  		$(R)  Create .env from .env.example with generated secrets."
-	@echo "  $(T)install      		$(R)  Install ALL dependencies (Prod + Dev)."
-	@echo "  $(T)install-hooks		$(R)  Install git pre-commit hooks."
-	@echo ""
-	@echo "$(G)Docker:$(R)"
-	@echo "  $(T)build        		$(R)  Build the docker image (production)."
-	@echo "  $(T)build-test   		$(R)  Build the docker image (with dev deps for testing)."
-	@echo "  $(T)down         		$(R)  Stop and remove containers."
-	@echo "  $(T)logs         		$(R)  Follow logs."
-	@echo "  $(T)prune        		$(R)  Clean Docker system."
-	@echo "  $(T)ps           		$(R)  Show container status."
-	@echo "  $(T)restart      		$(R)  Restart all containers."
-	@echo "  $(T)shell        		$(R)  Open shell in container."
-	@echo "  $(T)up           		$(R)  Start the agent."
-	@echo ""
-	@echo "$(G)Development:$(R)"
-	@echo "  $(T)add ...      		$(R)  Add dependency (use --group dev or --dev if needed)."
-	@echo "  $(T)clean        		$(R)  Clean caches."
-	@echo "  $(T)coverage     		$(R)  Run tests with coverage."
-	@echo "  $(T)format       		$(R)  Format code (Ruff)."
-	@echo "  $(T)lint         		$(R)  Lint code without fixing."
-	@echo "  $(T)test ...     		$(R)  Run tests (local with $(RUNNER))."
-	@echo "  $(T)update       		$(R)  Update dependencies."
-	@echo ""
-	@echo "$(G)Versioning:$(R)"
-	@echo "  $(T)major/minor/patch	$(R)  Bump version and push tag (triggers CI/CD)."
-
-init-dotenv:
-	@echo "$(T)🔑 Initializing .env file...$(R)"
-	@if [ -f .env ]; then \
-		echo "$(R)⚠️  .env already exists. Skipping.$(R)"; \
-		exit 0; \
-	fi
-	@if [ ! -f .env.example ]; then \
-		echo "$(R)❌ .env.example not found$(R)"; \
-		exit 1; \
-	fi
-	@if ! command -v openssl >/dev/null 2>&1; then \
-		echo "$(R)❌ openssl not found. Please install it first.$(R)"; \
-		exit 1; \
-	fi
-	@echo "$(T)  → Copying .env.example...$(R)"
-	@cp .env.example .env
-	@echo "$(T)  → Generating secrets...$(R)"
-	@sed -i.bak "s|JWT_SECRET=.*|JWT_SECRET=$$(openssl rand -base64 32)|" .env
-	@sed -i.bak "s|JWT_REFRESH_SECRET=.*|JWT_REFRESH_SECRET=$$(openssl rand -base64 32)|" .env
-	@sed -i.bak "s|CREDS_KEY=.*|CREDS_KEY=$$(openssl rand -hex 16)|" .env
-	@sed -i.bak "s|CREDS_IV=.*|CREDS_IV=$$(openssl rand -hex 8)|" .env
-	@sed -i.bak "s|MEILI_MASTER_KEY=.*|MEILI_MASTER_KEY=$$(openssl rand -base64 32)|" .env
-	@sed -i.bak "s|AGENT_BRAIN_API_KEY=.*|AGENT_BRAIN_API_KEY=$$(openssl rand -base64 24)|" .env
-	@rm -f .env.bak
-	@echo "$(G)✅ .env created with generated secrets!$(R)"
-	@echo "$(T)⚠️  Don't forget to add your API keys:$(R)"
-	@echo "   - OPENAI_API_KEY"
-	@echo "   - DEEPSEEK_API_KEY"
-	@echo "   - TMDB_API_KEY (optional)"
-
-install: check-runner
-	@echo "$(T)📦 Installing FULL environment ($(RUNNER))...$(R)"
-	$(RUNNER_INSTALL)
-	@echo "✅ Environment ready (Prod + Dev)."
-
-install-hooks: check-runner
-	@echo "$(T)🔧 Installing hooks...$(R)"
-	$(RUNNER_HOOKS)
-	@echo "✅ Hooks ready."
-
-lint: check-runner
-	@echo "$(T)🔍 Linting code...$(R)"
-	$(RUNNER_RUN) ruff check .
-
-logs: check-docker
-	@echo "$(T)📋 Following logs...$(R)"
-	$(COMPOSE_CMD) logs -f
-
-major: _check_branch
-	@echo "$(T)💥 Bumping major...$(R)"
-	SKIP=all $(BUMP_CMD) major
-	@$(MAKE) -s _push_tag
-
-minor: _check_branch
-	@echo "$(T)✨ Bumping minor...$(R)"
-	SKIP=all $(BUMP_CMD) minor
-	@$(MAKE) -s _push_tag
-
-patch: _check_branch
-	@echo "$(T)🚀 Bumping patch...$(R)"
-	SKIP=all $(BUMP_CMD) patch
-	@$(MAKE) -s _push_tag
-
-prune: check-docker
-	@echo "$(T)🗑️  Pruning Docker resources...$(R)"
-	docker system prune -af
-	@echo "✅ Docker cleaned."
-
-ps: check-docker
-	@echo "$(T)📋 Container status:$(R)"
-	@$(COMPOSE_CMD) ps
-
-restart: check-docker
-	@echo "$(T)🔄 Restarting containers...$(R)"
-	$(COMPOSE_CMD) restart
-	@echo "✅ Containers restarted."
-
-run: check-runner
-	$(RUNNER_RUN) $(ARGS)
-
-shell: check-docker
-	@echo "$(T)🐚 Opening shell in $(SERVICE_NAME)...$(R)"
-	$(COMPOSE_CMD) exec $(SERVICE_NAME) /bin/sh
-
-test: check-runner
-	@echo "$(T)🧪 Running tests...$(R)"
-	$(RUNNER_RUN) pytest $(ARGS)
-
-up: check-docker
-	@echo "$(T)🚀 Starting Agent Media...$(R)"
-	$(COMPOSE_CMD) up -d
-	@echo "✅ System is up."
-
-update: check-runner
-	@echo "$(T)🔄 Updating dependencies...$(R)"
-	$(RUNNER_UPDATE)
-	@echo "✅ All packages up to date."
-
-_check_branch:
-	@curr=$$(git rev-parse --abbrev-ref HEAD); \
-	if [ "$$curr" != "main" ]; then \
-	   echo "❌ Error: not on the main branch"; exit 1; \
-	fi
-
-_ci-image-name:
-	@echo "IMAGE_NAME=$(IMAGE_NAME)"
-
-_ci-run-tests: build-test
-	@echo "$(T)🧪 Running tests in Docker...$(R)"
 	docker run --rm \
 		-e DEEPSEEK_API_KEY \
 		-e TMDB_API_KEY \
+		-e QBITTORRENT_URL \
 		$(IMAGE_NAME):test pytest
-	@echo "✅ Tests passed."
+	@echo "✓ Tests passed."
 
-_push_tag:
+_check-main:
-	@echo "$(T)📦 Pushing tag...$(R)"
+	@test "$$(git rev-parse --abbrev-ref HEAD)" = "main" \
-	git push --tags
+		|| (echo "✗ ERROR: Not on main branch" && exit 1)
-	@echo "✅ Tag pushed. Check CI for build status."
+
+# --- Help ---
+help:
+	@echo "Cleverly Crafted Unawareness - Management Commands"
+	@echo ""
+	@echo "Usage: make [target] [p=profile1,profile2]"
+	@echo ""
+	@echo "Setup:"
+	@echo "  bootstrap   Generate .env.alfred, .env.librechat, .env.secrets and .env.make"
+	@echo ""
+	@echo "Docker:"
+	@echo "  up          Start containers (default profile: core)"
+	@echo "              Example: make up p=rag,meili"
+	@echo "  down        Stop all containers"
+	@echo "  restart     Restart containers (supports p=...)"
+	@echo "  logs        Follow logs (supports p=...)"
+	@echo "  ps          Status of containers"
+	@echo "  shell       Open bash in the core container"
+	@echo "  build       Build the production Docker image"
+	@echo ""
+	@echo "Dev & Quality:"
+	@echo "  setup       Bootstrap .env and security keys"
+	@echo "  install     Install dependencies via uv"
+	@echo "  test        Run pytest suite"
+	@echo "  coverage    Run tests and generate HTML report"
+	@echo "  lint/format Quality and style checks"
+	@echo ""
+	@echo "Release:"
+	@echo "  major|minor|patch  Bump version and push tags (main branch only)"

README.md

@@ -0,0 +1,433 @@
+# Alfred Media Organizer 🎬
+
+An AI-powered agent for managing your local media library with natural language. Search, download, and organize movies and TV shows effortlessly through a conversational interface.
+
+[![Python 3.14](https://img.shields.io/badge/python-3.14-blue.svg)](https://www.python.org/downloads/)
+[![uv](https://img.shields.io/badge/dependency%20manager-uv-purple)](https://github.com/astral-sh/uv)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![Code style: ruff](https://img.shields.io/badge/code%20style-ruff-000000.svg)](https://github.com/astral-sh/ruff)
+
+## ✨ Features
+
+- 🤖 **Natural Language Interface** — Talk to your media library in plain language
+- 🔍 **Smart Search** — Find movies and TV shows via TMDB with rich metadata
+- 📥 **Torrent Integration** — Search and download via qBittorrent
+- 🧠 **Contextual Memory** — Remembers your preferences and conversation history
+- 📁 **Auto-Organization** — Moves and renames media files, resolves destinations, handles subtitles
+- 🎞️ **Subtitle Pipeline** — Identifies, matches, and places subtitle tracks automatically
+- 🔄 **Workflow Engine** — YAML-defined multi-step workflows (e.g. `organize_media`)
+- 🌐 **OpenAI-Compatible API** — Works with any OpenAI-compatible client (LibreChat, OpenWebUI, etc.)
+- 🔒 **Secure by Default** — Auto-generated secrets and encrypted credentials
+
+## 🏗️ Architecture
+
+Built with **Domain-Driven Design (DDD)** principles for clean separation of concerns:
+
+```
+alfred/
+├── agent/              # AI agent orchestration
+│   ├── llm/            # LLM clients (Ollama, DeepSeek)
+│   ├── tools/          # Tool implementations (api, filesystem, language)
+│   └── workflows/      # YAML-defined multi-step workflows
+├── application/        # Use cases & DTOs
+│   ├── movies/         # Movie search
+│   ├── torrents/       # Torrent management
+│   └── filesystem/     # File operations (move, list, subtitles, seed links)
+├── domain/             # Business logic & entities
+│   ├── media/          # Release parsing
+│   ├── movies/         # Movie entities
+│   ├── tv_shows/       # TV show entities & value objects
+│   ├── subtitles/      # Subtitle scanner, services, knowledge base
+│   └── shared/         # Common value objects (ImdbId, FilePath, FileSize)
+└── infrastructure/     # External services & persistence
+    ├── api/            # External API clients (TMDB, qBittorrent, Knaben)
+    ├── filesystem/     # File manager (hard-link based, path-traversal safe)
+    ├── persistence/    # Three-tier memory (LTM/STM/Episodic) + JSON repositories
+    └── subtitle/       # Subtitle infrastructure
+```
+
+### Key flows
+
+**Agent execution:** `agent.step(user_input)` → LLM call → if tool_calls, execute each via registry → loop until no tool calls or `max_tool_iterations` → return final response.
+
+**Media organization workflow:**
+1. `resolve_destination` — Determines target folder/filename from release name
+2. `move_media` — Hard-links file to library, deletes source
+3. `manage_subtitles` — Scans, classifies, and places subtitle tracks
+4. `create_seed_links` — Hard-links library file back to torrents/ for continued seeding
+
+**Memory tiers:**
+- **LTM** (`data/memory/ltm.json`) — Persisted config, media library, watchlist
+- **STM** — Conversation history (capped at `MAX_HISTORY_MESSAGES`)
+- **Episodic** — Transient search results, active downloads, recent errors
+
+## 🚀 Quick Start
+
+### Prerequisites
+
+- **Python 3.14+**
+- **uv** (dependency manager)
+- **Docker & Docker Compose** (recommended for full stack)
+- **API Keys:**
+  - TMDB API key ([get one here](https://www.themoviedb.org/settings/api))
+  - Optional: DeepSeek or other LLM provider keys
+
+### Installation
+
+```bash
+# Clone the repository
+git clone https://github.com/francwa/alfred_media_organizer.git
+cd alfred_media_organizer
+
+# Install dependencies
+make install
+
+# Install pre-commit hooks
+make install-hooks
+
+# Bootstrap environment (generates .env with secure secrets)
+make bootstrap
+
+# Validate your .env against the schema
+make validate
+
+# Edit .env with your API keys
+nano .env
+```
+
+### Running with Docker (Recommended)
+
+```bash
+# Start all services (LibreChat + Alfred + MongoDB + Ollama)
+make up
+
+# Or start with specific profiles
+make up p=rag,meili      # Include RAG and Meilisearch
+make up p=qbittorrent    # Include qBittorrent
+make up p=full           # Everything
+
+# View logs
+make logs
+
+# Stop all services
+make down
+```
+
+The web interface will be available at **http://localhost:3080**
+
+### Running Locally (Development)
+
+```bash
+uv run uvicorn alfred.app:app --reload --port 8000
+```
+
+## ⚙️ Configuration
+
+### Settings system
+
+`settings.toml` is the single source of truth. The schema flows:
+
+```
+settings.toml → settings_schema.py → settings_bootstrap.py → .env + .env.make → settings.py
+```
+
+To add a setting: define it in `settings.toml`, run `make bootstrap`, then access via `settings.my_new_setting`.
+
+```bash
+# First time setup
+make bootstrap
+
+# Validate existing .env against schema
+make validate
+
+# Re-run after settings.toml changes (existing secrets preserved)
+make bootstrap
+```
+
+**Never commit `.env` or `.env.make`** — both are gitignored and auto-generated.
+
+### Key settings (.env)
+
+```bash
+# --- CORE ---
+MAX_HISTORY_MESSAGES=10
+MAX_TOOL_ITERATIONS=10
+
+# --- LLM ---
+DEFAULT_LLM_PROVIDER=local     # local (Ollama) | deepseek
+OLLAMA_BASE_URL=http://ollama:11434
+OLLAMA_MODEL=llama3.3:latest
+LLM_TEMPERATURE=0.2
+
+# --- API KEYS ---
+TMDB_API_KEY=your-tmdb-key     # Required for movie/show search
+DEEPSEEK_API_KEY=              # Optional
+
+# --- SECURITY (auto-generated) ---
+JWT_SECRET=<auto>
+CREDS_KEY=<auto>
+MONGO_PASSWORD=<auto>
+```
+
+## 🐳 Docker Services
+
+### Docker Profiles
+
+| Profile | Extra services | Use case |
+|---------|---------------|----------|
+| (default) | — | LibreChat + Alfred + MongoDB + Ollama |
+| `meili` | Meilisearch | Fast full-text search |
+| `rag` | RAG API + VectorDB (PostgreSQL) | Document retrieval |
+| `qbittorrent` | qBittorrent | Torrent downloads |
+| `full` | All of the above | Complete setup |
+
+```bash
+make up              # Start (default profile)
+make up p=full       # Start with all services
+make down            # Stop
+make restart         # Restart
+make logs            # Follow logs
+make ps              # Container status
+```
+
+## 🛠️ Available Tools
+
+| Tool | Description |
+|------|-------------|
+| `find_media_imdb_id` | Search for movies/TV shows on TMDB by title |
+| `find_torrent` | Search for torrents across multiple indexers |
+| `get_torrent_by_index` | Get detailed info about a specific result |
+| `add_torrent_by_index` | Download a torrent from search results |
+| `add_torrent_to_qbittorrent` | Add a torrent via magnet link directly |
+| `resolve_destination` | Compute the target library path for a release |
+| `move_media` | Hard-link a file to its library destination |
+| `manage_subtitles` | Scan, classify, and place subtitle tracks |
+| `create_seed_links` | Prepare torrent folder so qBittorrent keeps seeding |
+| `learn` | Teach Alfred a new pattern (release group, naming convention) |
+| `set_path_for_folder` | Configure folder paths |
+| `list_folder` | List contents of a configured folder |
+| `set_language` | Set preferred language for the session |
+
+## 💬 Usage Examples
+
+### Via Web Interface (LibreChat)
+
+Navigate to **http://localhost:3080** and start chatting:
+
+```
+You: Find Inception in 1080p
+Alfred: I found 3 torrents for Inception (2010):
+        1. Inception.2010.1080p.BluRay.x264 (150 seeders) - 2.1 GB
+        2. Inception.2010.1080p.WEB-DL.x265 (80 seeders) - 1.8 GB
+        3. Inception.2010.1080p.REMUX (45 seeders) - 25 GB
+
+You: Download the first one
+Alfred: ✓ Added to qBittorrent! Download started.
+
+You: Organize the Breaking Bad S01 download
+Alfred: ✓ Resolved destination: /tv_shows/Breaking.Bad/Season 01/
+        ✓ Moved 6 episode files
+        ✓ Placed 6 subtitle tracks (fr, en)
+        ✓ Seed links created in /torrents/
+```
+
+### Via API
+
+```bash
+# Health check
+curl http://localhost:8000/health
+
+# Chat (OpenAI-compatible)
+curl -X POST http://localhost:8000/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "alfred",
+    "messages": [{"role": "user", "content": "Find The Matrix 4K"}]
+  }'
+
+# List models
+curl http://localhost:8000/v1/models
+
+# View memory state
+curl http://localhost:8000/memory/state
+```
+
+Alfred is compatible with any OpenAI-compatible client. Point it at `http://localhost:8000/v1`, model `alfred`.
+
+## 🧠 Memory System
+
+Alfred uses a three-tier memory system:
+
+| Tier | Storage | Contents | Lifetime |
+|------|---------|----------|----------|
+| **LTM** | JSON file (`data/memory/ltm.json`) | Config, library, watchlist, learned patterns | Permanent |
+| **STM** | RAM | Conversation history (capped) | Session |
+| **Episodic** | RAM | Search results, active downloads, errors | Short-lived |
+
+## 🧪 Development
+
+### Running Tests
+
+```bash
+# Run full suite (parallel)
+make test
+
+# Run with coverage report
+make coverage
+
+# Run a single file
+uv run pytest tests/test_agent.py -v
+
+# Run a single class
+uv run pytest tests/test_agent.py::TestAgentInit -v
+
+# Skip slow tests
+uv run pytest -m "not slow"
+```
+
+### Test coverage
+
+The suite covers:
+- **Agent loop** — tool execution, history, max iterations, error handling
+- **Tool registry** — OpenAI schema format, parameter extraction
+- **Prompts** — system prompt building, tool inclusion
+- **Memory** — LTM/STM/Episodic operations, persistence
+- **Filesystem tools** — path traversal security, folder listing
+- **File manager** — hard-link, move, seed links (real filesystem, no mocks)
+- **Application use cases** — `resolve_destination`, `create_seed_links`, `list_folder`, `move_media`
+- **Domain** — TV show/movie entities, shared value objects (`ImdbId`, `FilePath`, `FileSize`), subtitle scanner
+- **Repositories** — JSON-backed movie, TV show, subtitle repos
+- **Bootstrap** — secret generation, idempotency, URI construction
+- **Workflows** — YAML loading, structure validation
+- **Configuration** — boundary validation for all settings
+
+### Code Quality
+
+```bash
+make lint        # Ruff check --fix
+make format      # Ruff format + check --fix
+```
+
+### Adding a New Tool
+
+1. Implement the function in `alfred/agent/tools/`:
+
+```python
+# alfred/agent/tools/api.py
+def my_new_tool(param: str) -> dict[str, Any]:
+    """Short description shown to the LLM to decide when to call this tool."""
+    memory = get_memory()
+    # ...
+    return {"status": "ok", "data": result}
+```
+
+2. Register it in `alfred/agent/registry.py`:
+
+```python
+tool_functions = [
+    # ... existing tools ...
+    api_tools.my_new_tool,
+]
+```
+
+The registry auto-generates the JSON schema from the function signature and docstring.
+
+### Adding a Workflow
+
+Create a YAML file in `alfred/agent/workflows/`:
+
+```yaml
+name: my_workflow
+description: What this workflow does
+steps:
+  - tool: resolve_destination
+    description: Find where the file should go
+  - tool: move_media
+    description: Move the file
+```
+
+Workflows are loaded automatically at startup.
+
+### Version Management
+
+```bash
+# Must be on main branch
+make patch    # 0.1.7 → 0.1.8
+make minor    # 0.1.7 → 0.2.0
+make major    # 0.1.7 → 1.0.0
+```
+
+## 📚 API Reference
+
+### Endpoints
+
+| Method | Path | Description |
+|--------|------|-------------|
+| `GET` | `/health` | Health check |
+| `GET` | `/v1/models` | List models (OpenAI-compatible) |
+| `POST` | `/v1/chat/completions` | Chat (OpenAI-compatible, streaming supported) |
+| `GET` | `/memory/state` | Full memory dump (debug) |
+| `POST` | `/memory/clear-session` | Clear STM + Episodic |
+| `GET` | `/memory/episodic/search-results` | Current search results |
+
+## 🔧 Troubleshooting
+
+### Agent doesn't respond
+
+1. Check API keys in `.env`
+2. Verify the LLM is running:
+   ```bash
+   docker logs alfred-ollama
+   docker exec alfred-ollama ollama list
+   ```
+3. Check Alfred logs: `docker logs alfred-core`
+
+### qBittorrent connection failed
+
+1. Verify qBittorrent is running: `docker ps | grep qbittorrent`
+2. Check credentials in `.env` (`QBITTORRENT_URL`, `QBITTORRENT_USERNAME`, `QBITTORRENT_PASSWORD`)
+
+### Memory not persisting
+
+1. Check `data/` directory is writable
+2. Verify volume mounts in `docker-compose.yaml`
+
+### Bootstrap fails
+
+```bash
+make validate    # Check what's wrong with .env
+make bootstrap   # Regenerate (preserves existing secrets)
+```
+
+### Tests failing
+
+```bash
+uv run pytest tests/test_failing.py -v --tb=long
+```
+
+## 🤝 Contributing
+
+1. Fork the repository
+2. Create a feature branch: `git checkout -b feat/my-feature`
+3. Make your changes + add tests
+4. Run `make test && make lint && make format`
+5. Commit with [Conventional Commits](https://www.conventionalcommits.org/): `feat:`, `fix:`, `docs:`, `refactor:`, `test:`, `chore:`, `infra:`
+6. Open a Pull Request
+
+## 📄 License
+
+MIT License — see [LICENSE](LICENSE) file for details.
+
+## 🙏 Acknowledgments
+
+- [LibreChat](https://github.com/danny-avila/LibreChat) — Chat interface
+- [Ollama](https://ollama.ai/) — Local LLM runtime
+- [DeepSeek](https://www.deepseek.com/) — LLM provider
+- [TMDB](https://www.themoviedb.org/) — Movie & TV database
+- [qBittorrent](https://www.qbittorrent.org/) — Torrent client
+- [FastAPI](https://fastapi.tiangolo.com/) — Web framework
+- [uv](https://github.com/astral-sh/uv) — Fast Python package manager
+
+---
+
+<p align="center">Made with ❤️ by <a href="https://github.com/francwa">Francwa</a></p>

alfred/agent/__init__.py

@@ -1,6 +1,7 @@
 """Agent module for media library management."""
 
+from alfred.settings import settings
+
 from .agent import Agent
-from .config import settings
 
 __all__ = ["Agent", "settings"]

alfred/agent/agent.py

@@ -5,9 +5,9 @@ import logging
 from collections.abc import AsyncGenerator
 from typing import Any
 
-from infrastructure.persistence import get_memory
+from alfred.infrastructure.persistence import get_memory
+from alfred.settings import settings
 
-from .config import settings
 from .prompts import PromptBuilder
 from .registry import Tool, make_tools
 
@@ -21,17 +21,20 @@ class Agent:
     Uses OpenAI-compatible tool calling API.
     """
 
-    def __init__(self, llm, max_tool_iterations: int = 5):
+    def __init__(self, settings, llm, max_tool_iterations: int = 5):
         """
         Initialize the agent.
 
         Args:
+            settings: Application settings instance
             llm: LLM client with complete() method
             max_tool_iterations: Maximum number of tool execution iterations
         """
+        self.settings = settings
         self.llm = llm
-        self.tools: dict[str, Tool] = make_tools()
+        self.tools: dict[str, Tool] = make_tools(settings)
         self.prompt_builder = PromptBuilder(self.tools)
+        self.settings = settings
         self.max_tool_iterations = max_tool_iterations
 
     def step(self, user_input: str) -> str:
@@ -78,7 +81,7 @@ class Agent:
         tools_spec = self.prompt_builder.build_tools_spec()
 
         # Tool execution loop
-        for _iteration in range(self.max_tool_iterations):
+        for _iteration in range(self.settings.max_tool_iterations):
             # Call LLM with tools
             llm_result = self.llm.complete(messages, tools=tools_spec)
 
@@ -230,7 +233,7 @@ class Agent:
         tools_spec = self.prompt_builder.build_tools_spec()
 
         # Tool execution loop
-        for _iteration in range(self.max_tool_iterations):
+        for _iteration in range(self.settings.max_tool_iterations):
             # Call LLM with tools
             llm_result = self.llm.complete(messages, tools=tools_spec)
 

alfred/agent/llm/deepseek.py

@@ -6,7 +6,8 @@ from typing import Any
 import requests
 from requests.exceptions import HTTPError, RequestException, Timeout
 
-from ..config import settings
+from alfred.settings import Settings, settings
+
 from .exceptions import LLMAPIError, LLMConfigurationError
 
 logger = logging.getLogger(__name__)
@@ -21,6 +22,7 @@ class DeepSeekClient:
         base_url: str | None = None,
         model: str | None = None,
         timeout: int | None = None,
+        settings: Settings | None = None,
     ):
         """
         Initialize DeepSeek client.
@@ -34,10 +36,10 @@ class DeepSeekClient:
         Raises:
             LLMConfigurationError: If API key is missing
         """
-        self.api_key = api_key or settings.deepseek_api_key
+        self.api_key = api_key or self.settings.deepseek_api_key
-        self.base_url = base_url or settings.deepseek_base_url
+        self.base_url = base_url or self.settings.deepseek_base_url
-        self.model = model or settings.model
+        self.model = model or self.settings.deepseek_model
-        self.timeout = timeout or settings.request_timeout
+        self.timeout = timeout or self.settings.request_timeout
 
         if not self.api_key:
             raise LLMConfigurationError(
@@ -94,7 +96,7 @@ class DeepSeekClient:
         payload = {
             "model": self.model,
             "messages": messages,
-            "temperature": settings.temperature,
+            "temperature": settings.llm_temperature,
         }
 
         # Add tools if provided

alfred/agent/llm/ollama.py

@@ -1,13 +1,13 @@
 """Ollama LLM client with robust error handling."""
 
 import logging
-import os
 from typing import Any
 
 import requests
 from requests.exceptions import HTTPError, RequestException, Timeout
 
-from ..config import settings
+from alfred.settings import Settings
+
 from .exceptions import LLMAPIError, LLMConfigurationError
 
 logger = logging.getLogger(__name__)
@@ -32,6 +32,7 @@ class OllamaClient:
         model: str | None = None,
         timeout: int | None = None,
         temperature: float | None = None,
+        settings: Settings | None = None,
     ):
         """
         Initialize Ollama client.
@@ -45,13 +46,11 @@ class OllamaClient:
         Raises:
             LLMConfigurationError: If configuration is invalid
         """
-        self.base_url = base_url or os.getenv(
+        self.base_url = base_url or settings.ollama_base_url
-            "OLLAMA_BASE_URL", "http://localhost:11434"
+        self.model = model or settings.ollama_model
-        )
-        self.model = model or os.getenv("OLLAMA_MODEL", "llama3.2")
         self.timeout = timeout or settings.request_timeout
         self.temperature = (
-            temperature if temperature is not None else settings.temperature
+            temperature if temperature is not None else settings.llm_temperature
         )
 
         if not self.base_url:

alfred/agent/prompts.py

@@ -3,7 +3,8 @@
 import json
 from typing import Any
 
-from infrastructure.persistence import get_memory
+from alfred.infrastructure.persistence import get_memory
+from alfred.infrastructure.persistence.memory import MemoryRegistry
 
 from .registry import Tool
 
@@ -13,6 +14,7 @@ class PromptBuilder:
 
     def __init__(self, tools: dict[str, Tool]):
         self.tools = tools
+        self._memory_registry = MemoryRegistry()
 
     def build_tools_spec(self) -> list[dict[str, Any]]:
         """Build the tool specification for the LLM API."""
@@ -52,7 +54,7 @@ class PromptBuilder:
             # Show first 5 results
             for i, result in enumerate(result_list[:5]):
                 name = result.get("name", "Unknown")
-                lines.append(f"  {i+1}. {name}")
+                lines.append(f"  {i + 1}. {name}")
             if len(result_list) > 5:
                 lines.append(f"  ... and {len(result_list) - 5} more")
 
@@ -109,11 +111,30 @@ class PromptBuilder:
 
         return "\n".join(lines)
 
+    def _format_memory_schema(self) -> str:
+        """Describe available memory components so the agent knows what to read/write and when."""
+        schema = self._memory_registry.schema()
+        tier_labels = {"ltm": "LONG-TERM (persisted)", "stm": "SHORT-TERM (session)", "episodic": "EPISODIC (volatile)"}
+        lines = ["MEMORY COMPONENTS:"]
+
+        for tier, components in schema.items():
+            if not components:
+                continue
+            lines.append(f"\n  [{tier_labels.get(tier, tier.upper())}]")
+            for c in components:
+                access = c.get("access", "read")
+                lines.append(f"  {c['name']} ({access}): {c['description']}")
+                for field_name, field_desc in c.get("fields", {}).items():
+                    lines.append(f"    · {field_name}: {field_desc}")
+
+        return "\n".join(lines)
+
     def _format_config_context(self, memory) -> str:
         """Format configuration context."""
         lines = ["CURRENT CONFIGURATION:"]
-        if memory.ltm.config:
+        folders = {**memory.ltm.workspace.as_dict(), **memory.ltm.library_paths.to_dict()}
-            for key, value in memory.ltm.config.items():
+        if folders:
+            for key, value in folders.items():
                 lines.append(f"  - {key}: {value}")
         else:
             lines.append("  (no configuration set)")
@@ -138,6 +159,9 @@ class PromptBuilder:
         tools_desc = self._format_tools_description()
         tools_section = f"\nAVAILABLE TOOLS:\n{tools_desc}" if tools_desc else ""
 
+        # Memory schema
+        memory_schema = self._format_memory_schema()
+
         # Configuration
         config_section = self._format_config_context(memory)
         if config_section:
@@ -172,6 +196,8 @@ EXAMPLES:
 
 {language_instruction}
 {tools_section}
+
+{memory_schema}
 {config_section}
 {stm_context}
 {episodic_context}

alfred/agent/registry.py

@@ -78,10 +78,13 @@ def _create_tool_from_function(func: Callable) -> Tool:
     )
 
 
-def make_tools() -> dict[str, Tool]:
+def make_tools(settings) -> dict[str, Tool]:
     """
     Create and register all available tools.
 
+    Args:
+        settings: Application settings instance
+
     Returns:
         Dictionary mapping tool names to Tool objects
     """
@@ -94,6 +97,11 @@ def make_tools() -> dict[str, Tool]:
     tool_functions = [
         fs_tools.set_path_for_folder,
         fs_tools.list_folder,
+        fs_tools.resolve_destination,
+        fs_tools.move_media,
+        fs_tools.manage_subtitles,
+        fs_tools.create_seed_links,
+        fs_tools.learn,
         api_tools.find_media_imdb_id,
         api_tools.find_torrent,
         api_tools.add_torrent_by_index,

alfred/agent/tools/api.py

@@ -3,12 +3,12 @@
 import logging
 from typing import Any
 
-from application.movies import SearchMovieUseCase
+from alfred.application.movies import SearchMovieUseCase
-from application.torrents import AddTorrentUseCase, SearchTorrentsUseCase
+from alfred.application.torrents import AddTorrentUseCase, SearchTorrentsUseCase
-from infrastructure.api.knaben import knaben_client
+from alfred.infrastructure.api.knaben import knaben_client
-from infrastructure.api.qbittorrent import qbittorrent_client
+from alfred.infrastructure.api.qbittorrent import qbittorrent_client
-from infrastructure.api.tmdb import tmdb_client
+from alfred.infrastructure.api.tmdb import tmdb_client
-from infrastructure.persistence import get_memory
+from alfred.infrastructure.persistence import get_memory
 
 logger = logging.getLogger(__name__)
 

alfred/agent/tools/filesystem.py

@@ -0,0 +1,230 @@
+"""Filesystem tools for folder management."""
+
+from pathlib import Path
+from typing import Any
+
+import alfred as _alfred_pkg
+import yaml
+
+from alfred.application.filesystem import (
+    CreateSeedLinksUseCase,
+    ListFolderUseCase,
+    ManageSubtitlesUseCase,
+    MoveMediaUseCase,
+    ResolveDestinationUseCase,
+    SetFolderPathUseCase,
+)
+from alfred.infrastructure.filesystem import FileManager
+
+_LEARNED_ROOT = Path(_alfred_pkg.__file__).parent.parent / "data" / "knowledge"
+
+
+def move_media(source: str, destination: str) -> dict[str, Any]:
+    """
+    Move a media file to a destination path.
+
+    Copies the file safely first (with integrity check), then deletes the source.
+    Use this to organise a downloaded file into the media library.
+
+    Args:
+        source: Absolute path to the source file.
+        destination: Absolute path to the destination file (must not already exist).
+
+    Returns:
+        Dict with status, source, destination, filename, and size — or error details.
+    """
+    file_manager = FileManager()
+    use_case = MoveMediaUseCase(file_manager)
+    return use_case.execute(source, destination).to_dict()
+
+
+def resolve_destination(
+    release_name: str,
+    source_file: str,
+    tmdb_title: str,
+    tmdb_year: int,
+    tmdb_episode_title: str | None = None,
+    confirmed_folder: str | None = None,
+) -> dict[str, Any]:
+    """
+    Compute the destination path in the media library for a release.
+
+    Call this before move_media to get the correct library path. Handles:
+    - Parsing the release name (quality, codec, group, season/episode)
+    - Looking up any existing series folder in the library
+    - Applying group-conflict rules (asks user if ambiguous)
+    - Building the full destination path with correct naming conventions
+
+    Args:
+        release_name: Raw release folder or file name
+                      (e.g. "Oz.S03.1080p.WEBRip.x265-KONTRAST").
+        source_file: Absolute path to the source video file (used for extension).
+        tmdb_title: Canonical show/movie title from TMDB (e.g. "Oz").
+        tmdb_year: Release/start year from TMDB (e.g. 1997).
+        tmdb_episode_title: Episode title from TMDB for single-episode releases
+                            (e.g. "The Routine"). Omit for season packs and movies.
+        confirmed_folder: If a previous call returned needs_clarification, pass
+                          the user-chosen folder name here to proceed.
+
+    Returns:
+        On success: dict with status, library_file, series_folder, season_folder,
+                    series_folder_name, season_folder_name, filename,
+                    is_new_series_folder.
+        On ambiguity: dict with status="needs_clarification", question, options.
+        On error: dict with status="error", error, message.
+    """
+    use_case = ResolveDestinationUseCase()
+    return use_case.execute(
+        release_name=release_name,
+        source_file=source_file,
+        tmdb_title=tmdb_title,
+        tmdb_year=tmdb_year,
+        tmdb_episode_title=tmdb_episode_title,
+        confirmed_folder=confirmed_folder,
+    ).to_dict()
+
+
+def create_seed_links(library_file: str, original_download_folder: str) -> dict[str, Any]:
+    """
+    Prepare a torrent subfolder so qBittorrent can keep seeding after a move.
+
+    Hard-links the video file from the library into torrents/<original_folder_name>/,
+    then copies all remaining files from the original download folder (subtitles,
+    .nfo, .jpg, .txt, …) so the torrent data is complete.
+
+    Call this after move_media when the user wants to keep seeding.
+
+    Args:
+        library_file: Absolute path to the video file now in the library.
+        original_download_folder: Absolute path to the original download folder
+            (may still contain subs, nfo, and other release files).
+
+    Returns:
+        Dict with status, torrent_subfolder, linked_file, copied_files,
+        copied_count, skipped — or error details.
+    """
+    file_manager = FileManager()
+    use_case = CreateSeedLinksUseCase(file_manager)
+    return use_case.execute(library_file, original_download_folder).to_dict()
+
+
+def manage_subtitles(source_video: str, destination_video: str) -> dict[str, Any]:
+    """
+    Place subtitle files alongside an organised video file.
+
+    Scans for subtitle files (.srt, .ass, .ssa, .vtt, .sub) next to the source
+    video, filters them according to the user's SubtitlePreferences (languages,
+    min size, SDH, forced), and hard-links the passing files next to the
+    destination video with the correct naming convention:
+        fr.srt / fr.sdh.srt / fr.forced.srt / en.srt …
+
+    Call this right after move_media or copy_media, passing the same source and
+    destination paths. If no subtitles are found, returns ok with placed_count=0.
+
+    Args:
+        source_video: Absolute path to the original video file (in the download folder).
+        destination_video: Absolute path to the placed video file (in the library).
+
+    Returns:
+        Dict with status, placed list (source, destination, filename), placed_count,
+        skipped_count — or error details.
+    """
+    file_manager = FileManager()
+    use_case = ManageSubtitlesUseCase(file_manager)
+    return use_case.execute(source_video, destination_video).to_dict()
+
+
+def learn(pack: str, category: str, key: str, values: list[str]) -> dict[str, Any]:
+    """
+    Teach Alfred a new token mapping and persist it to the learned knowledge pack.
+
+    Use this when a subtitle file contains an unrecognised token — after confirming
+    with the user what the token means, call learn() to persist it so Alfred
+    recognises it in future scans.
+
+    Args:
+        pack: Knowledge pack name. Currently only "subtitles" is supported.
+        category: Category within the pack: "languages", "types", or "formats".
+        key: The entry key — e.g. ISO 639-1 language code ("es"), type id ("sdh").
+        values: List of tokens to add — e.g. ["spanish", "espanol", "spa"].
+
+    Returns:
+        Dict with status, added_count, and the updated token list.
+    """
+    _VALID_PACKS = {"subtitles"}
+    _VALID_CATEGORIES = {"languages", "types", "formats"}
+
+    if pack not in _VALID_PACKS:
+        return {"status": "error", "error": "unknown_pack", "message": f"Unknown pack '{pack}'. Valid: {sorted(_VALID_PACKS)}"}
+
+    if category not in _VALID_CATEGORIES:
+        return {"status": "error", "error": "unknown_category", "message": f"Unknown category '{category}'. Valid: {sorted(_VALID_CATEGORIES)}"}
+
+    learned_path = _LEARNED_ROOT / "subtitles_learned.yaml"
+    _LEARNED_ROOT.mkdir(parents=True, exist_ok=True)
+
+    data: dict = {}
+    if learned_path.exists():
+        try:
+            with open(learned_path, encoding="utf-8") as f:
+                data = yaml.safe_load(f) or {}
+        except Exception as e:
+            return {"status": "error", "error": "read_failed", "message": str(e)}
+
+    cat_data = data.setdefault(category, {})
+    entry = cat_data.setdefault(key, {"tokens": []})
+    existing = entry.get("tokens", [])
+    new_tokens = [v for v in values if v not in existing]
+    entry["tokens"] = existing + new_tokens
+
+    tmp = learned_path.with_suffix(".yaml.tmp")
+    try:
+        with open(tmp, "w", encoding="utf-8") as f:
+            yaml.safe_dump(data, f, allow_unicode=True, default_flow_style=False, sort_keys=False)
+        tmp.rename(learned_path)
+    except Exception as e:
+        tmp.unlink(missing_ok=True)
+        return {"status": "error", "error": "write_failed", "message": str(e)}
+
+    return {
+        "status": "ok",
+        "pack": pack,
+        "category": category,
+        "key": key,
+        "added_count": len(new_tokens),
+        "tokens": entry["tokens"],
+    }
+
+
+def set_path_for_folder(folder_name: str, path_value: str) -> dict[str, Any]:
+    """
+    Set a folder path in the configuration.
+
+    Args:
+        folder_name: Name of folder to set (download, tvshow, movie, torrent).
+        path_value: Absolute path to the folder.
+
+    Returns:
+        Dict with status or error information.
+    """
+    file_manager = FileManager()
+    use_case = SetFolderPathUseCase(file_manager)
+    response = use_case.execute(folder_name, path_value)
+    return response.to_dict()
+
+
+def list_folder(folder_type: str, path: str = ".") -> dict[str, Any]:
+    """
+    List contents of a configured folder.
+
+    Args:
+        folder_type: Type of folder to list (download, tvshow, movie, torrent).
+        path: Relative path within the folder (default: root).
+
+    Returns:
+        Dict with folder contents or error information.
+    """
+    file_manager = FileManager()
+    use_case = ListFolderUseCase(file_manager)
+    response = use_case.execute(folder_type, path)
+    return response.to_dict()

alfred/agent/tools/language.py

@@ -3,7 +3,7 @@
 import logging
 from typing import Any
 
-from infrastructure.persistence import get_memory
+from alfred.infrastructure.persistence import get_memory
 
 logger = logging.getLogger(__name__)
 

alfred/agent/workflows/__init__.py

@@ -0,0 +1,3 @@
+from .loader import WorkflowLoader
+
+__all__ = ["WorkflowLoader"]

alfred/agent/workflows/loader.py

@@ -0,0 +1,52 @@
+"""WorkflowLoader — autodiscovers and loads workflow YAML files.
+
+Scans the workflows/ directory for all .yaml files and exposes them
+as dicts. No manual registration needed — drop a new .yaml file and
+it will be picked up automatically.
+"""
+
+import logging
+from pathlib import Path
+
+import yaml
+
+logger = logging.getLogger(__name__)
+
+_WORKFLOWS_DIR = Path(__file__).parent
+
+
+class WorkflowLoader:
+    """
+    Loads all workflow definitions from the workflows/ directory.
+
+    Usage:
+        loader = WorkflowLoader()
+        all_workflows = loader.all()
+        workflow = loader.get("organize_media")
+    """
+
+    def __init__(self):
+        self._workflows: dict[str, dict] = {}
+        self._load()
+
+    def _load(self) -> None:
+        for path in sorted(_WORKFLOWS_DIR.glob("*.yaml")):
+            try:
+                data = yaml.safe_load(path.read_text(encoding="utf-8"))
+                name = data.get("name") or path.stem
+                self._workflows[name] = data
+                logger.info(f"WorkflowLoader: Loaded '{name}' from {path.name}")
+            except Exception as e:
+                logger.warning(f"WorkflowLoader: Could not load {path.name}: {e}")
+
+    def all(self) -> dict[str, dict]:
+        """Return all loaded workflows keyed by name."""
+        return self._workflows
+
+    def get(self, name: str) -> dict | None:
+        """Return a specific workflow by name, or None if not found."""
+        return self._workflows.get(name)
+
+    def names(self) -> list[str]:
+        """Return all available workflow names."""
+        return list(self._workflows.keys())

alfred/agent/workflows/manage_subtitles.yaml

@@ -0,0 +1,69 @@
+name: manage_subtitles
+description: >
+  Place subtitle files alongside a video that has just been organised into the library.
+  Detects the release pattern automatically, identifies and classifies all tracks,
+  filters by user rules, and hard-links matching files to the destination.
+  If any tracks are unrecognised, asks the user and optionally teaches Alfred.
+
+trigger:
+  examples:
+    - "handle subtitles for The X-Files S01E01"
+    - "place the subs next to the file"
+    - "subtitles are in the Subs/ folder"
+    - "add subtitles"
+
+tools:
+  - manage_subtitles
+  - learn
+
+memory:
+  SubtitlePreferences: read
+  Workflow: read-write
+
+steps:
+  - id: place_subtitles
+    tool: manage_subtitles
+    description: >
+      Detect release pattern, identify and classify all subtitle tracks,
+      filter by rules, hard-link matching files next to the destination video.
+      Reads SubtitlePreferences from LTM for language/type/format filtering.
+    params:
+      source_video: "{source_video}"
+      destination_video: "{destination_video}"
+      imdb_id: "{imdb_id}"
+      media_type: "{media_type}"
+      release_group: "{release_group}"
+      season: "{season}"
+      episode: "{episode}"
+    on_result:
+      ok_placed_zero: skip               # no subtitles found — not an error
+      needs_clarification: ask_user      # unrecognised tokens found
+
+  - id: ask_user
+    description: >
+      Some tracks could not be classified. Show the user the unresolved tokens
+      and ask if they want to teach Alfred what they mean.
+      If yes → go to learn_tokens. If no → end workflow.
+    ask_user:
+      question: >
+        I could not identify some tokens in the subtitle files: {unresolved}.
+        Do you want to teach me what they mean?
+      answers:
+        yes: { next_step: learn_tokens }
+        no:  { next_step: end }
+
+  - id: learn_tokens
+    tool: learn
+    description: >
+      Persist a new token mapping to the learned knowledge pack so Alfred
+      recognises it in future scans without asking again.
+    params:
+      pack: "subtitles"
+      category: "{token_category}"   # "languages" or "types"
+      key: "{token_key}"             # e.g. "es", "de"
+      values: "{token_values}"       # e.g. ["spanish", "espanol"]
+
+subtitle_naming:
+  standard:  "{lang}.{ext}"
+  sdh:       "{lang}.sdh.{ext}"
+  forced:    "{lang}.forced.{ext}"

alfred/agent/workflows/organize_media.yaml

@@ -0,0 +1,82 @@
+name: organize_media
+description: >
+  Organise a downloaded series or movie into the media library.
+  Triggered when the user asks to move/organize a specific title.
+  Always moves the video file. Optionally creates seed links in the
+  torrents folder so qBittorrent can keep seeding.
+
+trigger:
+  examples:
+    - "organize Breaking Bad"
+    - "organise Severance season 2"
+    - "move Inception to my library"
+    - "organize Breaking Bad season 1, keep seeding"
+
+tools:
+  - list_folder
+  - find_media_imdb_id
+  - resolve_destination
+  - move_media
+  - manage_subtitles
+  - create_seed_links
+
+memory:
+  WorkspacePaths: read
+  LibraryPaths: read
+  Library: read-write
+  Workflow: read-write
+  Entities: read-write
+
+steps:
+  - id: list_downloads
+    tool: list_folder
+    description: List the download folder to find the target files.
+    params:
+      folder_type: download
+
+  - id: identify_media
+    tool: find_media_imdb_id
+    description: Confirm title, type (series/movie), and metadata via TMDB.
+
+  - id: resolve_destination
+    tool: resolve_destination
+    description: >
+      Compute the correct destination path in the library.
+      Uses the release name + TMDB metadata to build folder and file names.
+      If multiple series folders exist for this title, returns
+      needs_clarification and the user must pick one (re-call with confirmed_folder).
+
+  - id: move_file
+    tool: move_media
+    description: >
+      Move the video file to library_file returned by resolve_destination.
+
+  - id: handle_subtitles
+    tool: manage_subtitles
+    description: >
+      Place subtitle files alongside the video in the library.
+      Pass the original source path and the new library destination path.
+    on_missing: skip
+
+  - id: ask_seeding
+    ask_user:
+      question: "Do you want to keep seeding this torrent?"
+      answers:
+        "yes": { next_step: create_seed_links }
+        "no":  { next_step: update_library }
+
+  - id: create_seed_links
+    tool: create_seed_links
+    description: >
+      Hard-link the library video file back into torrents/<original_folder>/
+      and copy all remaining files from the original download folder
+      (subs, nfo, jpg, …) so the torrent stays complete for seeding.
+
+  - id: update_library
+    memory_write: Library
+    description: Add the entry to the LTM library after a successful move.
+
+naming_convention:
+  # Resolved by domain entities (Movie, Episode) — not hardcoded here
+  tv_show: "{title}/Season {season:02d}/{title}.S{season:02d}E{episode:02d}.{ext}"
+  movie: "{title} ({year})/{title}.{year}.{ext}"

alfred/app.py

@@ -2,22 +2,21 @@
 
 import json
 import logging
-import os
 import time
 import uuid
+from pathlib import Path
 from typing import Any
 
 from fastapi import FastAPI, HTTPException
 from fastapi.responses import JSONResponse, StreamingResponse
-from fastapi.staticfiles import StaticFiles
 from pydantic import BaseModel, Field, validator
 
-from agent.agent import Agent
+from alfred.agent.agent import Agent
-from agent.config import settings
+from alfred.agent.llm.deepseek import DeepSeekClient
-from agent.llm.deepseek import DeepSeekClient
+from alfred.agent.llm.exceptions import LLMAPIError, LLMConfigurationError
-from agent.llm.exceptions import LLMAPIError, LLMConfigurationError
+from alfred.agent.llm.ollama import OllamaClient
-from agent.llm.ollama import OllamaClient
+from alfred.infrastructure.persistence import get_memory, init_memory
-from infrastructure.persistence import get_memory, init_memory
+from alfred.settings import settings
 
 logging.basicConfig(
     level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s"
@@ -30,38 +29,33 @@ app = FastAPI(
     version="0.2.0",
 )
 
-# TODO: Make a variable
+memory_path = Path(settings.data_storage_dir) / "memory"
-manifests = "manifests"
+init_memory(storage_dir=str(memory_path))
-# Sécurité : on vérifie que le dossier existe pour ne pas faire planter l'app au démarrage
+logger.info(f"Memory context initialized (path: {memory_path})")
-if os.path.exists(manifests):
-    app.mount("/manifests", StaticFiles(directory=manifests), name="manifests")
-else:
-    print(
-        f"⚠️ ATTENTION : Le dossier '{manifests}' est introuvable. Le plugin ne marchera pas."
-    )
-
-# Initialize memory context at startup
-# Use /data/memory in Docker, fallback to memory_data for local dev
-storage_dir = os.getenv("MEMORY_STORAGE_DIR", "memory_data")
-init_memory(storage_dir=storage_dir)
-logger.info(f"Memory context initialized (storage: {storage_dir})")
 
 # Initialize LLM based on environment variable
-llm_provider = os.getenv("LLM_PROVIDER", "deepseek").lower()
+llm_provider = settings.default_llm_provider.lower()
+
 
 try:
-    if llm_provider == "ollama":
+    if llm_provider == "local":
-        logger.info("Using Ollama LLM")
+        logger.info("Using local Ollama LLM")
-        llm = OllamaClient()
+        llm = OllamaClient(settings=settings)
-    else:
+    elif llm_provider == "deepseek":
         logger.info("Using DeepSeek LLM")
         llm = DeepSeekClient()
+    elif llm_provider == "claude":
+        raise ValueError(f"LLM provider not fully implemented: {llm_provider}")
+    else:
+        raise ValueError(f"Unknown LLM provider: {llm_provider}")
 except LLMConfigurationError as e:
     logger.error(f"Failed to initialize LLM: {e}")
     raise
 
 # Initialize agent
-agent = Agent(llm=llm, max_tool_iterations=settings.max_tool_iterations)
+agent = Agent(
+    settings=settings, llm=llm, max_tool_iterations=settings.max_tool_iterations
+)
 logger.info("Agent Media API initialized")
 
 
@@ -116,7 +110,7 @@ def extract_last_user_content(messages: list[dict[str, Any]]) -> str:
 @app.get("/health")
 async def health_check():
     """Health check endpoint."""
-    return {"status": "healthy", "version": "0.2.0"}
+    return {"status": "healthy", "version": f"v{settings.alfred_version}"}
 
 
 @app.get("/v1/models")

alfred/application/filesystem/__init__.py

@@ -0,0 +1,32 @@
+"""Filesystem use cases."""
+
+from .create_seed_links import CreateSeedLinksUseCase
+from .dto import (
+    CreateSeedLinksResponse,
+    ListFolderResponse,
+    ManageSubtitlesResponse,
+    MoveMediaResponse,
+    PlacedSubtitle,
+    SetFolderPathResponse,
+)
+from .list_folder import ListFolderUseCase
+from .manage_subtitles import ManageSubtitlesUseCase
+from .move_media import MoveMediaUseCase
+from .resolve_destination import ResolveDestinationUseCase, ResolvedDestination
+from .set_folder_path import SetFolderPathUseCase
+
+__all__ = [
+    "SetFolderPathUseCase",
+    "ListFolderUseCase",
+    "CreateSeedLinksUseCase",
+    "MoveMediaUseCase",
+    "ManageSubtitlesUseCase",
+    "ResolveDestinationUseCase",
+    "ResolvedDestination",
+    "SetFolderPathResponse",
+    "ListFolderResponse",
+    "CreateSeedLinksResponse",
+    "MoveMediaResponse",
+    "ManageSubtitlesResponse",
+    "PlacedSubtitle",
+]

alfred/application/filesystem/create_seed_links.py

@@ -0,0 +1,54 @@
+"""CreateSeedLinksUseCase — prepares a torrent folder for continued seeding."""
+
+import logging
+
+from alfred.infrastructure.filesystem import FileManager
+from alfred.infrastructure.persistence import get_memory
+
+from .dto import CreateSeedLinksResponse
+
+logger = logging.getLogger(__name__)
+
+
+class CreateSeedLinksUseCase:
+    """
+    Prepares a torrent subfolder so qBittorrent can keep seeding after a move.
+
+    Hard-links the video file from the library back into torrents/<original_folder>/,
+    then copies all remaining files from the original download folder (subs, nfo, …).
+    """
+
+    def __init__(self, file_manager: FileManager):
+        self.file_manager = file_manager
+
+    def execute(
+        self, library_file: str, original_download_folder: str
+    ) -> CreateSeedLinksResponse:
+        memory = get_memory()
+        torrent_folder = memory.ltm.workspace.torrent
+
+        if not torrent_folder:
+            return CreateSeedLinksResponse(
+                status="error",
+                error="torrent_folder_not_set",
+                message="Torrent folder is not configured. Use set_path_for_folder to set it.",
+            )
+
+        result = self.file_manager.create_seed_links(
+            library_file, original_download_folder, torrent_folder
+        )
+
+        if result.get("status") == "ok":
+            return CreateSeedLinksResponse(
+                status="ok",
+                torrent_subfolder=result.get("torrent_subfolder"),
+                linked_file=result.get("linked_file"),
+                copied_files=result.get("copied_files"),
+                copied_count=result.get("copied_count", 0),
+                skipped=result.get("skipped"),
+            )
+        return CreateSeedLinksResponse(
+            status="error",
+            error=result.get("error"),
+            message=result.get("message"),
+        )

alfred/application/filesystem/dto.py

@@ -0,0 +1,209 @@
+"""Filesystem application DTOs."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+
+
+@dataclass
+class CopyMediaResponse:
+    """Response from copying a media file."""
+
+    status: str
+    source: str | None = None
+    destination: str | None = None
+    filename: str | None = None
+    size: int | None = None
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self) -> dict:
+        if self.error:
+            return {"status": self.status, "error": self.error, "message": self.message}
+        return {
+            "status": self.status,
+            "source": self.source,
+            "destination": self.destination,
+            "filename": self.filename,
+            "size": self.size,
+        }
+
+
+@dataclass
+class MoveMediaResponse:
+    """Response from moving a media file."""
+
+    status: str
+    source: str | None = None
+    destination: str | None = None
+    filename: str | None = None
+    size: int | None = None
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self) -> dict:
+        if self.error:
+            return {"status": self.status, "error": self.error, "message": self.message}
+        return {
+            "status": self.status,
+            "source": self.source,
+            "destination": self.destination,
+            "filename": self.filename,
+            "size": self.size,
+        }
+
+
+@dataclass
+class SetFolderPathResponse:
+    """Response from setting a folder path."""
+
+    status: str
+    folder_name: str | None = None
+    path: str | None = None
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self):
+        """Convert to dict for agent compatibility."""
+        result = {"status": self.status}
+
+        if self.error:
+            result["error"] = self.error
+            result["message"] = self.message
+        else:
+            if self.folder_name:
+                result["folder_name"] = self.folder_name
+            if self.path:
+                result["path"] = self.path
+
+        return result
+
+
+@dataclass
+class PlacedSubtitle:
+    """One subtitle file successfully placed."""
+
+    source: str
+    destination: str
+    filename: str
+
+    def to_dict(self) -> dict:
+        return {"source": self.source, "destination": self.destination, "filename": self.filename}
+
+
+@dataclass
+class UnresolvedTrack:
+    """A subtitle track that needs agent clarification before placement."""
+
+    raw_tokens: list[str]
+    file_path: str | None = None
+    file_size_kb: float | None = None
+    reason: str = ""   # "unknown_language" | "low_confidence"
+
+    def to_dict(self) -> dict:
+        return {
+            "raw_tokens": self.raw_tokens,
+            "file_path": self.file_path,
+            "file_size_kb": self.file_size_kb,
+            "reason": self.reason,
+        }
+
+
+@dataclass
+class AvailableSubtitle:
+    """One subtitle track available on an embedded media item."""
+
+    language: str        # ISO 639-2 code
+    subtitle_type: str   # "standard" | "sdh" | "forced" | "unknown"
+
+    def to_dict(self) -> dict:
+        return {"language": self.language, "type": self.subtitle_type}
+
+
+@dataclass
+class ManageSubtitlesResponse:
+    """Response from the manage_subtitles use case."""
+
+    status: str   # "ok" | "needs_clarification" | "error"
+    video_path: str | None = None
+    placed: list[PlacedSubtitle] | None = None
+    skipped_count: int = 0
+    unresolved: list[UnresolvedTrack] | None = None
+    available: list[AvailableSubtitle] | None = None   # embedded tracks summary
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self) -> dict:
+        if self.error:
+            return {"status": self.status, "error": self.error, "message": self.message}
+        result = {
+            "status": self.status,
+            "video_path": self.video_path,
+            "placed": [p.to_dict() for p in (self.placed or [])],
+            "placed_count": len(self.placed or []),
+            "skipped_count": self.skipped_count,
+        }
+        if self.unresolved:
+            result["unresolved"] = [u.to_dict() for u in self.unresolved]
+            result["unresolved_count"] = len(self.unresolved)
+        if self.available:
+            result["available"] = [a.to_dict() for a in self.available]
+        return result
+
+
+@dataclass
+class CreateSeedLinksResponse:
+    """Response from creating seed links for a torrent."""
+
+    status: str
+    torrent_subfolder: str | None = None
+    linked_file: str | None = None
+    copied_files: list[str] | None = None
+    copied_count: int = 0
+    skipped: list[str] | None = None
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self) -> dict:
+        if self.error:
+            return {"status": self.status, "error": self.error, "message": self.message}
+        return {
+            "status": self.status,
+            "torrent_subfolder": self.torrent_subfolder,
+            "linked_file": self.linked_file,
+            "copied_files": self.copied_files or [],
+            "copied_count": self.copied_count,
+            "skipped": self.skipped or [],
+        }
+
+
+@dataclass
+class ListFolderResponse:
+    """Response from listing a folder."""
+
+    status: str
+    folder_type: str | None = None
+    path: str | None = None
+    entries: list[str] | None = None
+    count: int | None = None
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self):
+        """Convert to dict for agent compatibility."""
+        result = {"status": self.status}
+
+        if self.error:
+            result["error"] = self.error
+            result["message"] = self.message
+        else:
+            if self.folder_type:
+                result["folder_type"] = self.folder_type
+            if self.path:
+                result["path"] = self.path
+            if self.entries is not None:
+                result["entries"] = self.entries
+            if self.count is not None:
+                result["count"] = self.count
+
+        return result

alfred/application/filesystem/list_folder.py

@@ -2,7 +2,7 @@
 
 import logging
 
-from infrastructure.filesystem import FileManager
+from alfred.infrastructure.filesystem import FileManager
 
 from .dto import ListFolderResponse
 

alfred/application/filesystem/manage_subtitles.py

@@ -0,0 +1,258 @@
+"""ManageSubtitlesUseCase — orchestrates the full subtitle pipeline for a video file."""
+
+import logging
+from pathlib import Path
+
+from alfred.domain.shared.value_objects import ImdbId
+from alfred.domain.subtitles.entities import SubtitleTrack
+from alfred.domain.subtitles.knowledge.base import SubtitleKnowledgeBase
+from alfred.domain.subtitles.knowledge.loader import KnowledgeLoader
+from alfred.domain.subtitles.services.identifier import SubtitleIdentifier
+from alfred.domain.subtitles.services.matcher import SubtitleMatcher
+from alfred.domain.subtitles.services.pattern_detector import PatternDetector
+from alfred.domain.subtitles.services.placer import PlacedTrack, SubtitlePlacer
+from alfred.domain.subtitles.services.utils import available_subtitles
+from alfred.domain.subtitles.value_objects import ScanStrategy
+from alfred.infrastructure.persistence.context import get_memory
+from alfred.infrastructure.subtitle.metadata_store import SubtitleMetadataStore
+from alfred.infrastructure.subtitle.rule_repository import RuleSetRepository
+
+from .dto import AvailableSubtitle, ManageSubtitlesResponse, PlacedSubtitle, UnresolvedTrack
+
+logger = logging.getLogger(__name__)
+
+
+def _infer_library_root(dest_video: Path, media_type: str) -> Path:
+    """
+    Infer the media library root folder from the destination video path.
+
+    TV show:  video → Season 01 → The X-Files   (3 levels up)
+    Movie:    video → Inception (2010)           (1 level up)
+    """
+    if media_type == "tv_show":
+        return dest_video.parent.parent
+    return dest_video.parent
+
+
+def _to_imdb_id(raw: str | None) -> ImdbId | None:
+    if not raw:
+        return None
+    try:
+        return ImdbId(raw)
+    except Exception:
+        return None
+
+
+class ManageSubtitlesUseCase:
+    """
+    Full subtitle pipeline:
+
+    1. Load knowledge base
+    2. Detect (or confirm) the release pattern
+    3. Identify all tracks (ffprobe + filesystem scan)
+    4. Load + resolve rules for this media
+    5. Match tracks against rules
+    6. If any tracks are unresolved → return needs_clarification (don't place yet)
+    7. Place matched tracks via hard-link
+    8. Persist to .alfred/metadata.yaml
+
+    The use case is stateless — all dependencies are instantiated inline.
+    """
+
+    def execute(
+        self,
+        source_video: str,
+        destination_video: str,
+        imdb_id: str | None = None,
+        media_type: str = "tv_show",
+        release_group: str | None = None,
+        season: int | None = None,
+        episode: int | None = None,
+        confirmed_pattern_id: str | None = None,
+    ) -> ManageSubtitlesResponse:
+        source_path = Path(source_video)
+        dest_path = Path(destination_video)
+
+        if not source_path.exists():
+            return ManageSubtitlesResponse(
+                status="error",
+                error="source_not_found",
+                message=f"Source video not found: {source_video}",
+            )
+
+        kb = SubtitleKnowledgeBase(KnowledgeLoader())
+        library_root = _infer_library_root(dest_path, media_type)
+        store = SubtitleMetadataStore(library_root)
+        repo = RuleSetRepository(library_root)
+
+        # --- Pattern resolution ---
+        pattern = self._resolve_pattern(
+            kb, store, source_path, confirmed_pattern_id, release_group
+        )
+        if pattern is None:
+            return ManageSubtitlesResponse(
+                status="error",
+                error="pattern_not_found",
+                message="Could not determine subtitle pattern for this release.",
+            )
+
+        # --- Identify ---
+        media_id = _to_imdb_id(imdb_id)
+        identifier = SubtitleIdentifier(kb)
+        metadata = identifier.identify(
+            video_path=source_path,
+            pattern=pattern,
+            media_id=media_id,
+            media_type=media_type,
+            release_group=release_group,
+        )
+
+        if metadata.total_count == 0:
+            logger.info(f"ManageSubtitles: no subtitle tracks found for {source_path.name}")
+            return ManageSubtitlesResponse(
+                status="ok",
+                video_path=destination_video,
+                placed=[],
+                skipped_count=0,
+            )
+
+        # --- Embedded short-circuit ---
+        if pattern.scan_strategy == ScanStrategy.EMBEDDED:
+            logger.info("ManageSubtitles: embedded pattern — skipping matcher")
+            available = [
+                AvailableSubtitle(
+                    language=t.language.code if t.language else "?",
+                    subtitle_type=t.subtitle_type.value,
+                )
+                for t in available_subtitles(metadata.embedded_tracks)
+            ]
+            return ManageSubtitlesResponse(
+                status="ok",
+                video_path=destination_video,
+                placed=[],
+                skipped_count=0,
+                available=available,
+            )
+
+        # --- Match (external only) ---
+        subtitle_prefs = None
+        try:
+            memory = get_memory()
+            subtitle_prefs = memory.ltm.subtitle_preferences
+        except Exception:
+            pass
+        rules = repo.load(release_group, subtitle_prefs).resolve()
+        matcher = SubtitleMatcher()
+        matched, unresolved = matcher.match(metadata.external_tracks, rules)
+
+        if unresolved:
+            logger.info(
+                f"ManageSubtitles: {len(unresolved)} unresolved track(s) — needs clarification"
+            )
+            return ManageSubtitlesResponse(
+                status="needs_clarification",
+                video_path=destination_video,
+                placed=[],
+                unresolved=[_to_unresolved_dto(t) for t in unresolved],
+            )
+
+        if not matched:
+            return ManageSubtitlesResponse(
+                status="ok",
+                video_path=destination_video,
+                placed=[],
+                skipped_count=metadata.total_count,
+            )
+
+        # --- Place ---
+        placer = SubtitlePlacer()
+        place_result = placer.place(matched, dest_path)
+
+        # --- Persist ---
+        if place_result.placed:
+            pairs = _pair_placed_with_tracks(place_result.placed, matched)
+            store.append_history(pairs, season, episode, release_group)
+
+        placed_dtos = [
+            PlacedSubtitle(
+                source=str(p.source),
+                destination=str(p.destination),
+                filename=p.filename,
+            )
+            for p in place_result.placed
+        ]
+
+        return ManageSubtitlesResponse(
+            status="ok",
+            video_path=destination_video,
+            placed=placed_dtos,
+            skipped_count=place_result.skipped_count,
+        )
+
+    def _resolve_pattern(
+        self,
+        kb: SubtitleKnowledgeBase,
+        store: SubtitleMetadataStore,
+        source_path: Path,
+        confirmed_pattern_id: str | None,
+        release_group: str | None,
+    ):
+        # 1. Explicit override from caller
+        if confirmed_pattern_id:
+            p = kb.pattern(confirmed_pattern_id)
+            if p:
+                return p
+            logger.warning(f"ManageSubtitles: unknown pattern '{confirmed_pattern_id}'")
+
+        # 2. Previously confirmed in metadata store
+        stored_id = store.confirmed_pattern()
+        if stored_id:
+            p = kb.pattern(stored_id)
+            if p:
+                logger.debug(f"ManageSubtitles: using confirmed pattern '{stored_id}'")
+                return p
+
+        # 3. Auto-detect
+        release_root = source_path.parent
+        detector = PatternDetector(kb)
+        result = detector.detect(release_root, source_path)
+
+        if result["detected"] and result["confidence"] >= 0.6:
+            logger.info(
+                f"ManageSubtitles: auto-detected pattern '{result['detected'].id}' "
+                f"(confidence={result['confidence']:.2f})"
+            )
+            return result["detected"]
+
+        # 4. Fallback — adjacent (safest default)
+        logger.info("ManageSubtitles: falling back to 'adjacent' pattern")
+        return kb.pattern("adjacent")
+
+
+def _to_unresolved_dto(track: SubtitleTrack, min_confidence: float = 0.7) -> UnresolvedTrack:
+    reason = "unknown_language" if track.language is None else "low_confidence"
+    return UnresolvedTrack(
+        raw_tokens=track.raw_tokens,
+        file_path=str(track.file_path) if track.file_path else None,
+        file_size_kb=track.file_size_kb,
+        reason=reason,
+    )
+
+
+def _pair_placed_with_tracks(
+    placed: list[PlacedTrack],
+    tracks: list[SubtitleTrack],
+) -> list[tuple[PlacedTrack, SubtitleTrack]]:
+    """
+    Pair each PlacedTrack with its originating SubtitleTrack by source path.
+    Falls back to positional matching if paths don't align.
+    """
+    track_by_path = {t.file_path: t for t in tracks if t.file_path}
+    pairs = []
+    for p in placed:
+        track = track_by_path.get(p.source)
+        if track is None and tracks:
+            track = tracks[0]   # positional fallback
+        if track:
+            pairs.append((p, track))
+    return pairs

alfred/application/filesystem/move_media.py

@@ -0,0 +1,43 @@
+"""Move media use case."""
+
+import logging
+
+from alfred.infrastructure.filesystem import FileManager
+
+from .dto import MoveMediaResponse
+
+logger = logging.getLogger(__name__)
+
+
+class MoveMediaUseCase:
+    """Use case for moving a media file to a destination (copy + delete source)."""
+
+    def __init__(self, file_manager: FileManager):
+        self.file_manager = file_manager
+
+    def execute(self, source: str, destination: str) -> MoveMediaResponse:
+        """
+        Move a media file from source to destination.
+
+        Args:
+            source: Absolute path to the source file.
+            destination: Absolute path to the destination file.
+
+        Returns:
+            MoveMediaResponse with success or error information.
+        """
+        result = self.file_manager.move_file(source, destination)
+
+        if result.get("status") == "ok":
+            return MoveMediaResponse(
+                status="ok",
+                source=result.get("source"),
+                destination=result.get("destination"),
+                filename=result.get("filename"),
+                size=result.get("size"),
+            )
+        return MoveMediaResponse(
+            status="error",
+            error=result.get("error"),
+            message=result.get("message"),
+        )

alfred/application/filesystem/resolve_destination.py

@@ -0,0 +1,246 @@
+"""
+ResolveDestinationUseCase — compute the library destination path for a release.
+
+Steps:
+1. Parse the release name
+2. Look up TMDB for title + year (+ episode title if single episode)
+3. Scan the library for an existing series folder
+4. Apply group-conflict rules
+5. Return the computed paths (or needs_clarification if ambiguous)
+"""
+
+from __future__ import annotations
+
+import logging
+import re
+from dataclasses import dataclass, field
+from pathlib import Path
+
+from alfred.domain.media.release_parser import ParsedRelease, parse_release
+from alfred.infrastructure.persistence import get_memory
+
+logger = logging.getLogger(__name__)
+
+# Characters forbidden on Windows filesystems (served via NFS)
+_WIN_FORBIDDEN = re.compile(r'[?:*"<>|\\]')
+
+
+def _sanitise(text: str) -> str:
+    return _WIN_FORBIDDEN.sub("", text)
+
+
+# ---------------------------------------------------------------------------
+# DTOs
+# ---------------------------------------------------------------------------
+
+@dataclass
+class ResolvedDestination:
+    """All computed paths for a release, ready to hand to move_media."""
+
+    status: str                         # "ok" | "needs_clarification" | "error"
+
+    # Populated on "ok"
+    library_file: str | None = None     # absolute path of the destination video file
+    series_folder: str | None = None    # absolute path of the series root folder
+    season_folder: str | None = None    # absolute path of the season subfolder
+    series_folder_name: str | None = None   # just the folder name (for display)
+    season_folder_name: str | None = None
+    filename: str | None = None
+    is_new_series_folder: bool = False  # True if we're creating the folder
+
+    # Populated on "needs_clarification"
+    question: str | None = None
+    options: list[str] | None = None    # existing group folder names to pick from
+
+    # Populated on "error"
+    error: str | None = None
+    message: str | None = None
+
+    def to_dict(self) -> dict:
+        if self.status == "error":
+            return {"status": self.status, "error": self.error, "message": self.message}
+        if self.status == "needs_clarification":
+            return {
+                "status": self.status,
+                "question": self.question,
+                "options": self.options or [],
+            }
+        return {
+            "status": self.status,
+            "library_file": self.library_file,
+            "series_folder": self.series_folder,
+            "season_folder": self.season_folder,
+            "series_folder_name": self.series_folder_name,
+            "season_folder_name": self.season_folder_name,
+            "filename": self.filename,
+            "is_new_series_folder": self.is_new_series_folder,
+        }
+
+
+# ---------------------------------------------------------------------------
+# Use case
+# ---------------------------------------------------------------------------
+
+class ResolveDestinationUseCase:
+    """
+    Compute the full destination path for a media file being organised.
+
+    The caller provides:
+    - release_name: the raw release folder/file name
+    - source_file: path to the actual video file (to get extension)
+    - tmdb_title: canonical title from TMDB
+    - tmdb_year: release year from TMDB
+    - tmdb_episode_title: episode title from TMDB (None for movies / season packs)
+    - confirmed_folder: if the user already answered needs_clarification, pass
+                        the chosen folder name here to skip the check
+
+    Returns a ResolvedDestination.
+    """
+
+    def execute(
+        self,
+        release_name: str,
+        source_file: str,
+        tmdb_title: str,
+        tmdb_year: int,
+        tmdb_episode_title: str | None = None,
+        confirmed_folder: str | None = None,
+    ) -> ResolvedDestination:
+        parsed = parse_release(release_name)
+        ext = Path(source_file).suffix  # ".mkv"
+
+        if parsed.is_movie:
+            return self._resolve_movie(parsed, tmdb_title, tmdb_year, ext)
+        return self._resolve_tvshow(
+            parsed, tmdb_title, tmdb_year, tmdb_episode_title, ext, confirmed_folder
+        )
+
+    # ------------------------------------------------------------------
+    # Movie
+    # ------------------------------------------------------------------
+
+    def _resolve_movie(
+        self, parsed: ParsedRelease, tmdb_title: str, tmdb_year: int, ext: str
+    ) -> ResolvedDestination:
+        memory = get_memory()
+        movies_root = memory.ltm.library_paths.get("movie")
+        if not movies_root:
+            return ResolvedDestination(
+                status="error",
+                error="library_not_set",
+                message="Movie library path is not configured.",
+            )
+
+        folder_name = _sanitise(parsed.movie_folder_name(tmdb_title, tmdb_year))
+        filename = _sanitise(parsed.movie_filename(tmdb_title, tmdb_year, ext))
+
+        folder_path = Path(movies_root) / folder_name
+        file_path = folder_path / filename
+
+        return ResolvedDestination(
+            status="ok",
+            library_file=str(file_path),
+            series_folder=str(folder_path),
+            series_folder_name=folder_name,
+            filename=filename,
+            is_new_series_folder=not folder_path.exists(),
+        )
+
+    # ------------------------------------------------------------------
+    # TV show
+    # ------------------------------------------------------------------
+
+    def _resolve_tvshow(
+        self,
+        parsed: ParsedRelease,
+        tmdb_title: str,
+        tmdb_year: int,
+        tmdb_episode_title: str | None,
+        ext: str,
+        confirmed_folder: str | None,
+    ) -> ResolvedDestination:
+        memory = get_memory()
+        tv_root = memory.ltm.library_paths.get("tv_show")
+        if not tv_root:
+            return ResolvedDestination(
+                status="error",
+                error="library_not_set",
+                message="TV show library path is not configured.",
+            )
+
+        tv_root_path = Path(tv_root)
+
+        # --- Find existing series folders for this title ---
+        existing = _find_existing_series_folders(tv_root_path, tmdb_title, tmdb_year)
+
+        # --- Determine series folder name ---
+        if confirmed_folder:
+            series_folder_name = confirmed_folder
+            is_new = not (tv_root_path / confirmed_folder).exists()
+        elif len(existing) == 0:
+            # No existing folder — create with release group
+            series_folder_name = _sanitise(parsed.show_folder_name(tmdb_title, tmdb_year))
+            is_new = True
+        elif len(existing) == 1:
+            # Exactly one match — use it regardless of group
+            series_folder_name = existing[0]
+            is_new = False
+        else:
+            # Multiple folders — ask user
+            return ResolvedDestination(
+                status="needs_clarification",
+                question=(
+                    f"Multiple folders found for '{tmdb_title}' in your library. "
+                    f"Which one should I use for this release ({parsed.group})?"
+                ),
+                options=existing,
+            )
+
+        # --- Build paths ---
+        season_folder_name = parsed.season_folder_name()
+        filename = _sanitise(
+            parsed.episode_filename(tmdb_episode_title, ext)
+            if not parsed.is_season_pack
+            else parsed.season_folder_name() + ext
+        )
+
+        series_path = tv_root_path / series_folder_name
+        season_path = series_path / season_folder_name
+        file_path = season_path / filename
+
+        return ResolvedDestination(
+            status="ok",
+            library_file=str(file_path),
+            series_folder=str(series_path),
+            season_folder=str(season_path),
+            series_folder_name=series_folder_name,
+            season_folder_name=season_folder_name,
+            filename=filename,
+            is_new_series_folder=is_new,
+        )
+
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+def _find_existing_series_folders(tv_root: Path, tmdb_title: str, tmdb_year: int) -> list[str]:
+    """
+    Return names of folders in tv_root that match the given title + year.
+
+    Matching is loose: normalised title (dots, no special chars) + year must
+    appear at the start of the folder name.
+    """
+    if not tv_root.exists():
+        return []
+
+    # Build a normalised prefix to match against: "Oz.1997"
+    clean_title = _sanitise(tmdb_title).replace(" ", ".")
+    prefix = f"{clean_title}.{tmdb_year}".lower()
+
+    matches = []
+    for entry in tv_root.iterdir():
+        if entry.is_dir() and entry.name.lower().startswith(prefix):
+            matches.append(entry.name)
+
+    return sorted(matches)

alfred/application/filesystem/set_folder_path.py

@@ -2,7 +2,7 @@
 
 import logging
 
-from infrastructure.filesystem import FileManager
+from alfred.infrastructure.filesystem import FileManager
 
 from .dto import SetFolderPathResponse
 

alfred/application/movies/search_movie.py

@@ -2,7 +2,7 @@
 
 import logging
 
-from infrastructure.api.tmdb import (
+from alfred.infrastructure.api.tmdb import (
     TMDBAPIError,
     TMDBClient,
     TMDBConfigurationError,

alfred/application/torrents/add_torrent.py

@@ -2,7 +2,7 @@
 
 import logging
 
-from infrastructure.api.qbittorrent import (
+from alfred.infrastructure.api.qbittorrent import (
     QBittorrentAPIError,
     QBittorrentAuthError,
     QBittorrentClient,

alfred/application/torrents/search_torrents.py

@@ -2,7 +2,11 @@
 
 import logging
 
-from infrastructure.api.knaben import KnabenAPIError, KnabenClient, KnabenNotFoundError
+from alfred.infrastructure.api.knaben import (
+    KnabenAPIError,
+    KnabenClient,
+    KnabenNotFoundError,
+)
 
 from .dto import SearchTorrentsResponse
 

alfred/domain/media/__init__.py

@@ -0,0 +1,5 @@
+"""Media domain — shared naming and release parsing."""
+
+from .release_parser import ParsedRelease, parse_release
+
+__all__ = ["ParsedRelease", "parse_release"]

alfred/domain/media/release_parser.py

@@ -0,0 +1,306 @@
+"""
+release_parser.py — Parse a release name into structured components.
+
+Handles both dot-separated and space-separated release names:
+    Oz.S03.1080p.WEBRip.x265-KONTRAST
+    Oz S03 1080p WEBRip x265-KONTRAST
+    Inception.2010.1080p.BluRay.x265-GROUP
+"""
+
+from __future__ import annotations
+
+import re
+from dataclasses import dataclass, field
+
+# Known quality tokens
+_QUALITIES = {"2160p", "1080p", "720p", "480p", "576p", "4k", "8k"}
+
+# Known source tokens (case-insensitive match)
+_SOURCES = {
+    "bluray", "blu-ray", "bdrip", "brrip",
+    "webrip", "web-rip", "webdl", "web-dl", "web",
+    "hdtv", "hdrip", "dvdrip", "dvd", "vodrip",
+    "amzn", "nf", "dsnp", "hmax", "atvp",
+}
+
+# Known codec tokens
+_CODECS = {
+    "x264", "x265", "h264", "h265", "hevc", "avc",
+    "xvid", "divx", "av1", "vp9",
+    "h.264", "h.265",
+}
+
+# Windows-forbidden characters (we strip these from display names)
+_WIN_FORBIDDEN = re.compile(r'[?:*"<>|\\]')
+
+# Episode/season pattern: S01, S01E02, S01E02E03, 1x02, etc.
+_SEASON_EP_RE = re.compile(
+    r"S(\d{1,2})(?:E(\d{2})(?:E(\d{2}))?)?",
+    re.IGNORECASE,
+)
+
+# Year pattern
+_YEAR_RE = re.compile(r"\b(19\d{2}|20\d{2})\b")
+
+
+@dataclass
+class ParsedRelease:
+    """Structured representation of a parsed release name."""
+
+    raw: str                        # original release name (untouched)
+    normalised: str                 # dots instead of spaces
+    title: str                      # show/movie title (dots, no year/season/tech)
+    year: int | None                # movie year or show start year (from TMDB)
+    season: int | None              # season number (None for movies)
+    episode: int | None             # first episode number (None if season-pack)
+    episode_end: int | None         # last episode for multi-ep (None otherwise)
+    quality: str | None             # 1080p, 2160p, …
+    source: str | None              # WEBRip, BluRay, …
+    codec: str | None               # x265, HEVC, …
+    group: str                      # release group, "UNKNOWN" if missing
+    tech_string: str                # quality.source.codec joined with dots
+
+    # -------------------------------------------------------------------------
+    # Derived helpers
+    # -------------------------------------------------------------------------
+
+    @property
+    def is_movie(self) -> bool:
+        return self.season is None
+
+    @property
+    def is_season_pack(self) -> bool:
+        return self.season is not None and self.episode is None
+
+    def show_folder_name(self, tmdb_title: str, tmdb_year: int) -> str:
+        """
+        Build the series root folder name.
+
+        Format: {Title}.{Year}.{Tech}-{Group}
+        Example: Oz.1997.1080p.WEBRip.x265-KONTRAST
+        """
+        title_part = _sanitise_for_fs(tmdb_title).replace(" ", ".")
+        tech = self.tech_string or "Unknown"
+        return f"{title_part}.{tmdb_year}.{tech}-{self.group}"
+
+    def season_folder_name(self) -> str:
+        """
+        Build the season subfolder name = normalised release name (no episode).
+
+        Example: Oz.S03.1080p.WEBRip.x265-KONTRAST
+        For a single-episode release we still strip the episode token so the
+        folder can hold the whole season.
+        """
+        return _strip_episode_from_normalised(self.normalised)
+
+    def episode_filename(self, tmdb_episode_title: str | None, ext: str) -> str:
+        """
+        Build the episode filename.
+
+        Format: {Title}.{SxxExx}.{EpisodeTitle}.{Tech}-{Group}.{ext}
+        Example: Oz.S01E01.The.Routine.1080p.WEBRip.x265-KONTRAST.mkv
+
+        If tmdb_episode_title is None, omits the episode title segment.
+        """
+        title_part = _sanitise_for_fs(self.title)   # already dotted from normalised
+        s = f"S{self.season:02d}" if self.season is not None else ""
+        e = f"E{self.episode:02d}" if self.episode is not None else ""
+        se = s + e
+
+        ep_title = ""
+        if tmdb_episode_title:
+            ep_title = "." + _sanitise_for_fs(tmdb_episode_title).replace(" ", ".")
+
+        tech = self.tech_string or "Unknown"
+        ext_clean = ext.lstrip(".")
+        return f"{title_part}.{se}{ep_title}.{tech}-{self.group}.{ext_clean}"
+
+    def movie_folder_name(self, tmdb_title: str, tmdb_year: int) -> str:
+        """
+        Build the movie folder name.
+
+        Format: {Title}.{Year}.{Tech}-{Group}
+        Example: Inception.2010.1080p.BluRay.x265-GROUP
+        """
+        return self.show_folder_name(tmdb_title, tmdb_year)
+
+    def movie_filename(self, tmdb_title: str, tmdb_year: int, ext: str) -> str:
+        """
+        Build the movie filename (same as folder name + extension).
+
+        Example: Inception.2010.1080p.BluRay.x265-GROUP.mkv
+        """
+        ext_clean = ext.lstrip(".")
+        return f"{self.movie_folder_name(tmdb_title, tmdb_year)}.{ext_clean}"
+
+
+# ---------------------------------------------------------------------------
+# Public API
+# ---------------------------------------------------------------------------
+
+def parse_release(name: str) -> ParsedRelease:
+    """
+    Parse a release name and return a ParsedRelease.
+
+    Accepts both dot-separated and space-separated names.
+    """
+    normalised = _normalise(name)
+    tokens = normalised.split(".")
+
+    season, episode, episode_end = _extract_season_episode(tokens)
+    quality, source, codec, group, tech_tokens = _extract_tech(tokens)
+    title = _extract_title(tokens, season, episode, tech_tokens)
+    year = _extract_year(tokens, title)
+
+    tech_parts = [p for p in [quality, source, codec] if p]
+    tech_string = ".".join(tech_parts)
+
+    return ParsedRelease(
+        raw=name,
+        normalised=normalised,
+        title=title,
+        year=year,
+        season=season,
+        episode=episode,
+        episode_end=episode_end,
+        quality=quality,
+        source=source,
+        codec=codec,
+        group=group,
+        tech_string=tech_string,
+    )
+
+
+# ---------------------------------------------------------------------------
+# Internal helpers
+# ---------------------------------------------------------------------------
+
+def _normalise(name: str) -> str:
+    """Replace spaces with dots, collapse multiple dots."""
+    s = name.replace(" ", ".")
+    s = re.sub(r"\.{2,}", ".", s)
+    return s.strip(".")
+
+
+def _sanitise_for_fs(text: str) -> str:
+    """Remove Windows-forbidden characters from a string."""
+    return _WIN_FORBIDDEN.sub("", text)
+
+
+def _extract_season_episode(tokens: list[str]) -> tuple[int | None, int | None, int | None]:
+    joined = ".".join(tokens)
+    m = _SEASON_EP_RE.search(joined)
+    if not m:
+        return None, None, None
+    season = int(m.group(1))
+    episode = int(m.group(2)) if m.group(2) else None
+    episode_end = int(m.group(3)) if m.group(3) else None
+    return season, episode, episode_end
+
+
+def _extract_tech(
+    tokens: list[str],
+) -> tuple[str | None, str | None, str | None, str, set[str]]:
+    """
+    Extract quality, source, codec, group from tokens.
+
+    Returns (quality, source, codec, group, tech_token_set).
+
+    Group extraction strategy (in priority order):
+    1. Token where prefix is a known codec: x265-GROUP
+    2. Last token in the list that contains a dash (fallback for 10bit-GROUP, AAC5.1-GROUP, etc.)
+    """
+    quality: str | None = None
+    source: str | None = None
+    codec: str | None = None
+    group = "UNKNOWN"
+    tech_tokens: set[str] = set()
+
+    for tok in tokens:
+        tl = tok.lower()
+
+        if tl in _QUALITIES:
+            quality = tok
+            tech_tokens.add(tok)
+            continue
+
+        if tl in _SOURCES:
+            source = tok
+            tech_tokens.add(tok)
+            continue
+
+        if "-" in tok:
+            parts = tok.rsplit("-", 1)
+            # codec-GROUP (highest priority for group)
+            if parts[0].lower() in _CODECS:
+                codec = parts[0]
+                group = parts[1] if parts[1] else "UNKNOWN"
+                tech_tokens.add(tok)
+                continue
+            # source with dash: Web-DL, WEB-DL, etc.
+            if parts[0].lower() in _SOURCES or tok.lower().replace("-", "") in _SOURCES:
+                source = tok
+                tech_tokens.add(tok)
+                continue
+
+        if tl in _CODECS:
+            codec = tok
+            tech_tokens.add(tok)
+
+    # Fallback: if group still UNKNOWN, use the rightmost token with a dash
+    # that isn't a known source (handles "10bit-Protozoan", "AAC5.1-YTS", etc.)
+    if group == "UNKNOWN":
+        for tok in reversed(tokens):
+            if "-" in tok:
+                parts = tok.rsplit("-", 1)
+                tl = tok.lower()
+                if tl in _SOURCES or tok.lower().replace("-", "") in _SOURCES:
+                    continue
+                if parts[1]:  # non-empty group part
+                    group = parts[1]
+                    break
+
+    return quality, source, codec, group, tech_tokens
+
+
+def _extract_title(tokens: list[str], season: int | None, episode: int | None, tech_tokens: set[str]) -> str:
+    """
+    Extract the title portion: everything before the first season/year/tech token.
+    """
+    title_parts = []
+    for tok in tokens:
+        # Stop at season token
+        if _SEASON_EP_RE.match(tok):
+            break
+        # Stop at year
+        if _YEAR_RE.fullmatch(tok):
+            break
+        # Stop at tech tokens
+        if tok in tech_tokens or tok.lower() in _QUALITIES | _SOURCES | _CODECS:
+            break
+        # Stop if token contains a dash (likely codec-GROUP)
+        if "-" in tok and any(p.lower() in _CODECS | _SOURCES for p in tok.split("-")):
+            break
+        title_parts.append(tok)
+
+    return ".".join(title_parts) if title_parts else tokens[0]
+
+
+def _extract_year(tokens: list[str], title: str) -> int | None:
+    """Extract a 4-digit year from tokens (only after the title)."""
+    title_len = len(title.split("."))
+    for tok in tokens[title_len:]:
+        m = _YEAR_RE.fullmatch(tok)
+        if m:
+            return int(m.group(1))
+    return None
+
+
+def _strip_episode_from_normalised(normalised: str) -> str:
+    """
+    Remove all episode parts (Exx) from a normalised release name, keeping Sxx.
+
+    Oz.S03E01.1080p...             → Oz.S03.1080p...
+    Archer.S14E09E10E11.1080p...   → Archer.S14.1080p...
+    """
+    return re.sub(r"(S\d{2})(E\d{2})+", r"\1", normalised, flags=re.IGNORECASE)

alfred/domain/subtitles/__init__.py

@@ -0,0 +1,37 @@
+"""Subtitles domain — subtitle identification, classification and placement."""
+
+from .aggregates import SubtitleRuleSet
+from .entities import MediaSubtitleMetadata, SubtitleTrack
+from .exceptions import SubtitleNotFound
+from .knowledge import KnowledgeLoader, SubtitleKnowledgeBase
+from .services import PatternDetector, SubtitleIdentifier, SubtitleMatcher
+from .value_objects import (
+    RuleScope,
+    ScanStrategy,
+    SubtitleFormat,
+    SubtitleLanguage,
+    SubtitleMatchingRules,
+    SubtitlePattern,
+    SubtitleType,
+    TypeDetectionMethod,
+)
+
+__all__ = [
+    "SubtitleTrack",
+    "MediaSubtitleMetadata",
+    "SubtitleRuleSet",
+    "SubtitleKnowledgeBase",
+    "KnowledgeLoader",
+    "SubtitleIdentifier",
+    "SubtitleMatcher",
+    "PatternDetector",
+    "SubtitleFormat",
+    "SubtitleLanguage",
+    "SubtitlePattern",
+    "SubtitleType",
+    "ScanStrategy",
+    "TypeDetectionMethod",
+    "SubtitleMatchingRules",
+    "RuleScope",
+    "SubtitleNotFound",
+]

alfred/domain/subtitles/aggregates.py

@@ -0,0 +1,90 @@
+"""Subtitle domain aggregates."""
+
+from dataclasses import dataclass, field
+from typing import Any
+
+from ..shared.value_objects import ImdbId
+from .knowledge.base import SubtitleKnowledgeBase
+from .value_objects import RuleScope, SubtitleMatchingRules
+
+
+def DEFAULT_RULES() -> SubtitleMatchingRules:
+    """Load default matching rules from subtitles.yaml (defaults section)."""
+    return SubtitleKnowledgeBase().default_rules()
+
+
+@dataclass
+class SubtitleRuleSet:
+    """
+    Rules for subtitle selection at a given scope level, with inheritance.
+
+    Only delta fields are stored — None means "inherit from parent".
+    Resolution order: global → release_group → show/movie → season → episode.
+
+    A RuleSet can also be pinned to a specific media item (imdb_id),
+    bypassing the scope hierarchy for that item.
+    """
+
+    scope: RuleScope
+    parent: "SubtitleRuleSet | None" = None
+    pinned_to: ImdbId | None = None
+
+    # Deltas — None = inherit
+    _languages: list[str] | None = field(default=None, repr=False)
+    _formats: list[str] | None = field(default=None, repr=False)
+    _types: list[str] | None = field(default=None, repr=False)
+    _format_priority: list[str] | None = field(default=None, repr=False)
+    _min_confidence: float | None = field(default=None, repr=False)
+
+    def resolve(self) -> SubtitleMatchingRules:
+        """
+        Walk the parent chain and merge deltas into effective rules.
+        Falls back to DEFAULT_RULES at the top of the chain.
+        """
+        base = self.parent.resolve() if self.parent else DEFAULT_RULES()
+        return SubtitleMatchingRules(
+            preferred_languages=self._languages or base.preferred_languages,
+            preferred_formats=self._formats or base.preferred_formats,
+            allowed_types=self._types or base.allowed_types,
+            format_priority=self._format_priority or base.format_priority,
+            min_confidence=self._min_confidence if self._min_confidence is not None else base.min_confidence,
+        )
+
+    def override(
+        self,
+        languages: list[str] | None = None,
+        formats: list[str] | None = None,
+        types: list[str] | None = None,
+        format_priority: list[str] | None = None,
+        min_confidence: float | None = None,
+    ) -> None:
+        """Set delta overrides at this scope level."""
+        if languages is not None:
+            self._languages = languages
+        if formats is not None:
+            self._formats = formats
+        if types is not None:
+            self._types = types
+        if format_priority is not None:
+            self._format_priority = format_priority
+        if min_confidence is not None:
+            self._min_confidence = min_confidence
+
+    def to_dict(self) -> dict:
+        """Serialize deltas only (for persistence in rules.yaml)."""
+        delta: dict[str, Any] = {}
+        if self._languages is not None:
+            delta["languages"] = self._languages
+        if self._formats is not None:
+            delta["formats"] = self._formats
+        if self._types is not None:
+            delta["types"] = self._types
+        if self._format_priority is not None:
+            delta["format_priority"] = self._format_priority
+        if self._min_confidence is not None:
+            delta["min_confidence"] = self._min_confidence
+        return {"scope": {"level": self.scope.level, "identifier": self.scope.identifier}, "override": delta}
+
+    @classmethod
+    def global_default(cls) -> "SubtitleRuleSet":
+        return cls(scope=RuleScope(level="global"))

alfred/domain/subtitles/entities.py

@@ -0,0 +1,87 @@
+"""Subtitle domain entities."""
+
+from dataclasses import dataclass, field
+from pathlib import Path
+
+from ..shared.value_objects import ImdbId
+from .value_objects import SubtitleFormat, SubtitleLanguage, SubtitleMatchingRules, SubtitleType
+
+
+@dataclass
+class SubtitleTrack:
+    """
+    A single subtitle track — either an external file or an embedded stream.
+
+    State can evolve: unknown → resolved after user clarification.
+    confidence reflects how certain we are about language + type classification.
+    """
+
+    # Classification (may be None if not yet resolved)
+    language: SubtitleLanguage | None
+    format: SubtitleFormat | None
+    subtitle_type: SubtitleType = SubtitleType.UNKNOWN
+
+    # Source
+    is_embedded: bool = False
+    file_path: Path | None = None       # None if embedded
+    file_size_kb: float | None = None
+    entry_count: int | None = None      # number of subtitle cues in the file
+
+    # Matching state
+    confidence: float = 0.0             # 0.0 → 1.0, not applicable for embedded
+    raw_tokens: list[str] = field(default_factory=list)   # tokens extracted from filename
+
+    def is_resolved(self) -> bool:
+        return self.language is not None
+
+    @property
+    def destination_name(self) -> str:
+        """
+        Compute the output filename per naming convention:
+            {lang}.{ext}
+            {lang}.sdh.{ext}
+            {lang}.forced.{ext}
+        """
+        if not self.language or not self.format:
+            raise ValueError("Cannot compute destination_name: language or format missing")
+        ext = self.format.extensions[0].lstrip(".")
+        parts = [self.language.code]
+        if self.subtitle_type == SubtitleType.SDH:
+            parts.append("sdh")
+        elif self.subtitle_type == SubtitleType.FORCED:
+            parts.append("forced")
+        return ".".join(parts) + "." + ext
+
+    def __repr__(self) -> str:
+        lang = self.language.code if self.language else "?"
+        fmt = self.format.id if self.format else "?"
+        src = "embedded" if self.is_embedded else str(self.file_path.name if self.file_path else "?")
+        return f"SubtitleTrack({lang}, {self.subtitle_type.value}, {fmt}, src={src}, conf={self.confidence:.2f})"
+
+
+@dataclass
+class MediaSubtitleMetadata:
+    """
+    Snapshot of all subtitle information known for a given media item.
+    Populated by the identifier service (ffprobe + filesystem scan).
+    """
+
+    media_id: ImdbId | None
+    media_type: str                         # "movie" | "tv_show"
+    embedded_tracks: list[SubtitleTrack] = field(default_factory=list)
+    external_tracks: list[SubtitleTrack] = field(default_factory=list)
+    release_group: str | None = None
+    detected_pattern_id: str | None = None  # pattern id from knowledge base
+    pattern_confirmed: bool = False
+
+    @property
+    def all_tracks(self) -> list[SubtitleTrack]:
+        return self.embedded_tracks + self.external_tracks
+
+    @property
+    def total_count(self) -> int:
+        return len(self.embedded_tracks) + len(self.external_tracks)
+
+    @property
+    def unresolved_tracks(self) -> list[SubtitleTrack]:
+        return [t for t in self.external_tracks if t.language is None]

alfred/domain/subtitles/knowledge/__init__.py

@@ -0,0 +1,4 @@
+from .base import SubtitleKnowledgeBase
+from .loader import KnowledgeLoader
+
+__all__ = ["SubtitleKnowledgeBase", "KnowledgeLoader"]

alfred/domain/subtitles/knowledge/base.py

@@ -0,0 +1,151 @@
+"""SubtitleKnowledgeBase — parsed, typed view of the loaded knowledge."""
+
+import logging
+from functools import cached_property
+
+from ..value_objects import (
+    ScanStrategy,
+    SubtitleFormat,
+    SubtitleLanguage,
+    SubtitleMatchingRules,
+    SubtitlePattern,
+    SubtitleType,
+    TypeDetectionMethod,
+)
+from .loader import KnowledgeLoader
+
+logger = logging.getLogger(__name__)
+
+
+class SubtitleKnowledgeBase:
+    """
+    Typed access to subtitle knowledge (formats, types, languages, patterns).
+
+    Built from KnowledgeLoader — call kb.reload() to pick up newly learned entries
+    without restarting.
+    """
+
+    def __init__(self, loader: KnowledgeLoader | None = None):
+        self._loader = loader or KnowledgeLoader()
+        self._build()
+
+    def _build(self) -> None:
+        data = self._loader.subtitles()
+
+        self._formats: dict[str, SubtitleFormat] = {}
+        for fid, fdata in data.get("formats", {}).items():
+            self._formats[fid] = SubtitleFormat(
+                id=fid,
+                extensions=fdata.get("extensions", []),
+                description=fdata.get("description", ""),
+            )
+
+        self._languages: dict[str, SubtitleLanguage] = {}
+        for code, ldata in data.get("languages", {}).items():
+            self._languages[code] = SubtitleLanguage(
+                code=code,
+                tokens=ldata.get("tokens", []),
+            )
+
+        # Build reverse token → language code map
+        self._lang_token_map: dict[str, str] = {}
+        for code, lang in self._languages.items():
+            for token in lang.tokens:
+                self._lang_token_map[token.lower()] = code
+
+        # Build reverse token → type map
+        self._type_token_map: dict[str, SubtitleType] = {}
+        for type_id, tdata in data.get("types", {}).items():
+            stype = SubtitleType(type_id)
+            for token in tdata.get("tokens", []):
+                self._type_token_map[token.lower()] = stype
+
+        d = data.get("defaults", {})
+        self._default_rules = SubtitleMatchingRules(
+            preferred_languages=d.get("languages", ["fra", "eng"]),
+            preferred_formats=d.get("formats", ["srt"]),
+            allowed_types=d.get("types", ["standard", "forced"]),
+            format_priority=d.get("format_priority", ["srt", "ass"]),
+            min_confidence=d.get("min_confidence", 0.7),
+        )
+
+        self._patterns: dict[str, SubtitlePattern] = {}
+        for pid, pdata in self._loader.patterns().items():
+            try:
+                self._patterns[pid] = SubtitlePattern(
+                    id=pid,
+                    description=pdata.get("description", ""),
+                    scan_strategy=ScanStrategy(pdata.get("scan_strategy", "adjacent")),
+                    root_folder=pdata.get("root_folder"),
+                    type_detection=TypeDetectionMethod(
+                        pdata.get("type_detection", {}).get("method", "token_in_name")
+                    ),
+                    version=pdata.get("version", "1.0"),
+                )
+            except ValueError as e:
+                logger.warning(f"SubtitleKnowledgeBase: skipping pattern '{pid}': {e}")
+
+    def reload(self) -> None:
+        self._loader = KnowledgeLoader()
+        self._build()
+        logger.info("SubtitleKnowledgeBase: reloaded")
+
+    # --- Defaults ---
+
+    def default_rules(self) -> SubtitleMatchingRules:
+        return self._default_rules
+
+    # --- Formats ---
+
+    def formats(self) -> dict[str, SubtitleFormat]:
+        return self._formats
+
+    def format_for_extension(self, ext: str) -> SubtitleFormat | None:
+        for fmt in self._formats.values():
+            if fmt.matches_extension(ext):
+                return fmt
+        return None
+
+    def known_extensions(self) -> set[str]:
+        exts = set()
+        for fmt in self._formats.values():
+            exts.update(fmt.extensions)
+        return exts
+
+    # --- Languages ---
+
+    def languages(self) -> dict[str, SubtitleLanguage]:
+        return self._languages
+
+    def language_for_token(self, token: str) -> SubtitleLanguage | None:
+        code = self._lang_token_map.get(token.lower())
+        return self._languages.get(code) if code else None
+
+    def is_known_lang_token(self, token: str) -> bool:
+        return token.lower() in self._lang_token_map
+
+    # --- Types ---
+
+    def type_for_token(self, token: str) -> SubtitleType | None:
+        return self._type_token_map.get(token.lower())
+
+    def is_known_type_token(self, token: str) -> bool:
+        return token.lower() in self._type_token_map
+
+    # --- Patterns ---
+
+    def patterns(self) -> dict[str, SubtitlePattern]:
+        return self._patterns
+
+    def pattern(self, pattern_id: str) -> SubtitlePattern | None:
+        return self._patterns.get(pattern_id)
+
+    def patterns_for_group(self, group_name: str) -> list[SubtitlePattern]:
+        group = self._loader.release_group(group_name)
+        if not group:
+            return []
+        return [
+            self._patterns[pid]
+            for pid in group.get("known_patterns", [])
+            if pid in self._patterns
+        ]

alfred/domain/subtitles/knowledge/loader.py

@@ -0,0 +1,131 @@
+"""KnowledgeLoader — autodiscovers and merges builtin + learned YAML knowledge packs."""
+
+import logging
+from pathlib import Path
+
+import yaml
+
+logger = logging.getLogger(__name__)
+
+import alfred as _alfred_pkg
+
+# Builtin knowledge — anchored on the alfred package itself, not on this file's depth
+_BUILTIN_ROOT = Path(_alfred_pkg.__file__).parent / "knowledge"
+
+# Learned knowledge — local to this instance, gitignored
+_LEARNED_ROOT = Path(_alfred_pkg.__file__).parent.parent / "data" / "knowledge"
+
+
+def _load_yaml(path: Path) -> dict:
+    try:
+        with open(path, encoding="utf-8") as f:
+            return yaml.safe_load(f) or {}
+    except FileNotFoundError:
+        return {}
+    except Exception as e:
+        logger.warning(f"KnowledgeLoader: could not load {path}: {e}")
+        return {}
+
+
+def _merge(base: dict, override: dict) -> dict:
+    """
+    Deep merge override into base.
+    Lists are extended (not replaced) — learned tokens are additive.
+    Scalar values in override win over base.
+    """
+    result = dict(base)
+    for key, val in override.items():
+        if key in result and isinstance(result[key], dict) and isinstance(val, dict):
+            result[key] = _merge(result[key], val)
+        elif key in result and isinstance(result[key], list) and isinstance(val, list):
+            # Extend list, deduplicate, preserve order
+            combined = result[key] + [v for v in val if v not in result[key]]
+            result[key] = combined
+        else:
+            result[key] = val
+    return result
+
+
+class KnowledgeLoader:
+    """
+    Loads subtitle knowledge from YAML files.
+
+    Builtin packs live in alfred/knowledge/ (versioned).
+    Learned packs live in data/knowledge/ (gitignored, instance-local).
+
+    Learned entries are merged additively — they can only add tokens/patterns,
+    never remove builtin ones.
+
+    Usage:
+        loader = KnowledgeLoader()
+        subtitles = loader.subtitles()   # merged subtitles.yaml
+        patterns = loader.patterns()     # all patterns, keyed by id
+        groups = loader.release_groups() # all release groups, keyed by name
+    """
+
+    def __init__(self):
+        self._cache: dict[str, dict] = {}
+        self._load()
+
+    def _load(self) -> None:
+        # Main subtitles knowledge
+        builtin = _load_yaml(_BUILTIN_ROOT / "subtitles.yaml")
+        learned = _load_yaml(_LEARNED_ROOT / "subtitles_learned.yaml")
+        self._cache["subtitles"] = _merge(builtin, learned)
+
+        # Patterns
+        self._cache["patterns"] = {}
+        for path in sorted((_BUILTIN_ROOT / "patterns").glob("*.yaml")):
+            data = _load_yaml(path)
+            pid = data.get("id", path.stem)
+            self._cache["patterns"][pid] = data
+
+        for path in sorted((_LEARNED_ROOT / "patterns").glob("*.yaml")):
+            data = _load_yaml(path)
+            pid = data.get("id", path.stem)
+            if pid in self._cache["patterns"]:
+                self._cache["patterns"][pid] = _merge(self._cache["patterns"][pid], data)
+            else:
+                self._cache["patterns"][pid] = data
+                logger.info(f"KnowledgeLoader: learned new pattern '{pid}'")
+
+        # Release groups
+        self._cache["release_groups"] = {}
+        for path in sorted((_BUILTIN_ROOT / "release_groups").glob("*.yaml")):
+            data = _load_yaml(path)
+            name = data.get("name", path.stem)
+            self._cache["release_groups"][name] = data
+
+        for path in sorted((_LEARNED_ROOT / "release_groups").glob("*.yaml")):
+            data = _load_yaml(path)
+            name = data.get("name", path.stem)
+            if name in self._cache["release_groups"]:
+                self._cache["release_groups"][name] = _merge(self._cache["release_groups"][name], data)
+            else:
+                self._cache["release_groups"][name] = data
+                logger.info(f"KnowledgeLoader: learned new release group '{name}'")
+
+        logger.info(
+            f"KnowledgeLoader: {len(self._cache['patterns'])} patterns, "
+            f"{len(self._cache['release_groups'])} release groups loaded"
+        )
+
+    def subtitles(self) -> dict:
+        return self._cache["subtitles"]
+
+    def patterns(self) -> dict[str, dict]:
+        return self._cache["patterns"]
+
+    def pattern(self, pattern_id: str) -> dict | None:
+        return self._cache["patterns"].get(pattern_id)
+
+    def release_groups(self) -> dict[str, dict]:
+        return self._cache["release_groups"]
+
+    def release_group(self, name: str) -> dict | None:
+        """Case-insensitive lookup."""
+        name_lower = name.lower()
+        for key, val in self._cache["release_groups"].items():
+            if key.lower() == name_lower:
+                return val
+        return None

alfred/domain/subtitles/scanner.py

@@ -0,0 +1,221 @@
+"""SubtitleScanner — inspects local subtitle files and filters them per user preferences.
+
+Given a video file path, the scanner:
+  1. Looks for subtitle files in the same directory as the video.
+  2. Optionally also inspects a Subs/ subfolder adjacent to the video.
+  3. Classifies each file (language, SDH, forced) from its filename.
+  4. Filters according to SubtitlePreferences (languages, min_size_kb, keep_sdh, keep_forced).
+  5. Returns a list of SubtitleCandidate — one per file that passes the filter,
+     with the destination filename already computed.
+
+Filename classification heuristics
+-----------------------------------
+We parse the stem of each subtitle file looking for known patterns:
+
+  fr.srt              → lang=fr, sdh=False, forced=False
+  fr.sdh.srt          → lang=fr, sdh=True
+  fr.hi.srt           → lang=fr, sdh=True  (hi = hearing-impaired, alias for sdh)
+  fr.forced.srt       → lang=fr, forced=True
+  Breaking.Bad.S01E01.French.srt  → lang=fr (keyword match)
+  Breaking.Bad.S01E01.VOSTFR.srt  → lang=fr (VOSTFR = French forced/foreign subs)
+
+Output naming convention (matches SubtitlePreferences docstring):
+  {lang}.srt
+  {lang}.sdh.srt
+  {lang}.forced.srt
+"""
+
+import logging
+from dataclasses import dataclass, field
+from pathlib import Path
+
+logger = logging.getLogger(__name__)
+
+# Subtitle file extensions we handle
+SUBTITLE_EXTENSIONS = {".srt", ".ass", ".ssa", ".vtt", ".sub"}
+
+# Language keyword map: lowercase token → ISO 639-1 code
+_LANG_KEYWORDS: dict[str, str] = {
+    # French
+    "fr": "fr",
+    "fra": "fr",
+    "french": "fr",
+    "francais": "fr",
+    "français": "fr",
+    "vf": "fr",
+    "vff": "fr",
+    "vostfr": "fr",
+    # English
+    "en": "en",
+    "eng": "en",
+    "english": "en",
+    # Spanish
+    "es": "es",
+    "spa": "es",
+    "spanish": "es",
+    "espanol": "es",
+    # German
+    "de": "de",
+    "deu": "de",
+    "ger": "de",
+    "german": "de",
+    # Italian
+    "it": "it",
+    "ita": "it",
+    "italian": "it",
+    # Portuguese
+    "pt": "pt",
+    "por": "pt",
+    "portuguese": "pt",
+    # Dutch
+    "nl": "nl",
+    "nld": "nl",
+    "dutch": "nl",
+    # Japanese
+    "ja": "ja",
+    "jpn": "ja",
+    "japanese": "ja",
+}
+
+# Tokens that indicate SDH / hearing-impaired
+_SDH_TOKENS = {"sdh", "hi", "hearing", "impaired", "cc", "closedcaption"}
+
+# Tokens that indicate forced subtitles
+_FORCED_TOKENS = {"forced", "foreign"}
+
+
+@dataclass
+class SubtitleCandidate:
+    """A subtitle file that passed the filter, ready to be placed."""
+
+    source_path: Path
+    language: str          # ISO 639-1 code, e.g. "fr"
+    is_sdh: bool
+    is_forced: bool
+    extension: str         # e.g. ".srt"
+
+    @property
+    def destination_name(self) -> str:
+        """
+        Compute the destination filename per naming convention:
+            {lang}.srt
+            {lang}.sdh.srt
+            {lang}.forced.srt
+        """
+        ext = self.extension.lstrip(".")
+        parts = [self.language]
+        if self.is_sdh:
+            parts.append("sdh")
+        elif self.is_forced:
+            parts.append("forced")
+        return ".".join(parts) + "." + ext
+
+
+def _classify(path: Path) -> tuple[str | None, bool, bool]:
+    """
+    Parse a subtitle filename and return (language_code, is_sdh, is_forced).
+
+    Returns (None, False, False) if the language cannot be determined.
+    """
+    stem = path.stem.lower()
+    # Split on dots, spaces, underscores, hyphens
+    import re
+    tokens = re.split(r"[\.\s_\-]+", stem)
+
+    language: str | None = None
+    is_sdh = False
+    is_forced = False
+
+    for token in tokens:
+        if token in _LANG_KEYWORDS:
+            language = _LANG_KEYWORDS[token]
+        if token in _SDH_TOKENS:
+            is_sdh = True
+        if token in _FORCED_TOKENS:
+            is_forced = True
+
+    return language, is_sdh, is_forced
+
+
+class SubtitleScanner:
+    """
+    Scans subtitle files next to a video and filters them per SubtitlePreferences.
+
+    Usage:
+        scanner = SubtitleScanner(prefs)
+        candidates = scanner.scan(video_path)
+        # Each candidate has .source_path and .destination_name
+    """
+
+    def __init__(self, languages: list[str], min_size_kb: int, keep_sdh: bool, keep_forced: bool):
+        self.languages = [l.lower() for l in languages]
+        self.min_size_kb = min_size_kb
+        self.keep_sdh = keep_sdh
+        self.keep_forced = keep_forced
+
+    def scan(self, video_path: Path) -> list[SubtitleCandidate]:
+        """
+        Return all subtitle candidates found next to the video that pass the filter.
+
+        Scans:
+          - Same directory as the video (flat siblings)
+          - Subs/ subfolder if present
+        """
+        candidates: list[SubtitleCandidate] = []
+        search_dirs = [video_path.parent]
+
+        subs_dir = video_path.parent / "Subs"
+        if subs_dir.is_dir():
+            search_dirs.append(subs_dir)
+            logger.debug(f"SubtitleScanner: found Subs/ folder at {subs_dir}")
+
+        for directory in search_dirs:
+            for path in sorted(directory.iterdir()):
+                if not path.is_file():
+                    continue
+                if path.suffix.lower() not in SUBTITLE_EXTENSIONS:
+                    continue
+
+                candidate = self._evaluate(path)
+                if candidate is not None:
+                    candidates.append(candidate)
+
+        logger.info(f"SubtitleScanner: {len(candidates)} candidate(s) found for {video_path.name}")
+        return candidates
+
+    def _evaluate(self, path: Path) -> SubtitleCandidate | None:
+        """Apply all filters to a single subtitle file. Returns None if it should be dropped."""
+        # Size filter
+        size_kb = path.stat().st_size / 1024
+        if size_kb < self.min_size_kb:
+            logger.debug(f"SubtitleScanner: skip {path.name} (too small: {size_kb:.1f} KB)")
+            return None
+
+        language, is_sdh, is_forced = _classify(path)
+
+        # Language filter
+        if language is None:
+            logger.debug(f"SubtitleScanner: skip {path.name} (language unknown)")
+            return None
+
+        if language not in self.languages:
+            logger.debug(f"SubtitleScanner: skip {path.name} (language '{language}' not in prefs)")
+            return None
+
+        # SDH filter
+        if is_sdh and not self.keep_sdh:
+            logger.debug(f"SubtitleScanner: skip {path.name} (SDH not wanted)")
+            return None
+
+        # Forced filter
+        if is_forced and not self.keep_forced:
+            logger.debug(f"SubtitleScanner: skip {path.name} (forced not wanted)")
+            return None
+
+        return SubtitleCandidate(
+            source_path=path,
+            language=language,
+            is_sdh=is_sdh,
+            is_forced=is_forced,
+            extension=path.suffix.lower(),
+        )

alfred/domain/subtitles/services/__init__.py

@@ -0,0 +1,13 @@
+from .identifier import SubtitleIdentifier
+from .matcher import SubtitleMatcher
+from .pattern_detector import PatternDetector
+from .placer import PlacedTrack, PlaceResult, SubtitlePlacer
+
+__all__ = [
+    "SubtitleIdentifier",
+    "SubtitleMatcher",
+    "PatternDetector",
+    "SubtitlePlacer",
+    "PlacedTrack",
+    "PlaceResult",
+]

alfred/domain/subtitles/services/identifier.py

@@ -0,0 +1,287 @@
+"""SubtitleIdentifier — finds and classifies all subtitle tracks for a video file."""
+
+import logging
+import re
+import subprocess
+import json
+from pathlib import Path
+
+from ...shared.value_objects import ImdbId
+from ..entities import MediaSubtitleMetadata, SubtitleTrack
+from ..knowledge.base import SubtitleKnowledgeBase
+from ..value_objects import ScanStrategy, SubtitlePattern, SubtitleType
+
+logger = logging.getLogger(__name__)
+
+
+def _tokenize(name: str) -> list[str]:
+    """Split a filename stem into lowercase tokens."""
+    return [t.lower() for t in re.split(r"[\.\s_\-]+", name) if t]
+
+
+def _count_entries(path: Path) -> int:
+    """Return the entry count of an SRT file by finding the last cue number."""
+    try:
+        with open(path, encoding="utf-8", errors="replace") as f:
+            lines = f.read().splitlines()
+        for line in reversed(lines):
+            if line.strip().isdigit():
+                return int(line.strip())
+        return 0
+    except Exception:
+        return 0
+
+
+class SubtitleIdentifier:
+    """
+    Finds all subtitle tracks for a given video file using a known pattern,
+    then attempts to classify each track (language, type, format).
+
+    Returns a MediaSubtitleMetadata with embedded + external tracks.
+    External tracks with unknown language or low confidence are left as-is —
+    the caller (use case) decides whether to ask the user for clarification.
+    """
+
+    def __init__(self, kb: SubtitleKnowledgeBase):
+        self.kb = kb
+
+    def identify(
+        self,
+        video_path: Path,
+        pattern: SubtitlePattern,
+        media_id: ImdbId | None,
+        media_type: str,
+        release_group: str | None = None,
+    ) -> MediaSubtitleMetadata:
+        metadata = MediaSubtitleMetadata(
+            media_id=media_id,
+            media_type=media_type,
+            release_group=release_group,
+            detected_pattern_id=pattern.id,
+        )
+
+        if pattern.scan_strategy == ScanStrategy.EMBEDDED:
+            metadata.embedded_tracks = self._scan_embedded(video_path)
+        else:
+            metadata.external_tracks = self._scan_external(video_path, pattern)
+            # Always also check for embedded tracks
+            metadata.embedded_tracks = self._scan_embedded(video_path)
+
+        return metadata
+
+    # ------------------------------------------------------------------
+    # Embedded tracks — ffprobe
+    # ------------------------------------------------------------------
+
+    def _scan_embedded(self, video_path: Path) -> list[SubtitleTrack]:
+        if not video_path.exists():
+            return []
+        try:
+            result = subprocess.run(
+                [
+                    "ffprobe", "-v", "quiet",
+                    "-print_format", "json",
+                    "-show_streams",
+                    "-select_streams", "s",
+                    str(video_path),
+                ],
+                capture_output=True, text=True, timeout=30,
+            )
+            data = json.loads(result.stdout)
+        except (subprocess.TimeoutExpired, json.JSONDecodeError, FileNotFoundError) as e:
+            logger.debug(f"SubtitleIdentifier: ffprobe failed for {video_path.name}: {e}")
+            return []
+
+        tracks = []
+        for stream in data.get("streams", []):
+            tags = stream.get("tags", {})
+            disposition = stream.get("disposition", {})
+            lang_code = tags.get("language", "")
+            title = tags.get("title", "")
+
+            lang = self.kb.language_for_token(lang_code) if lang_code else None
+
+            if disposition.get("hearing_impaired"):
+                stype = SubtitleType.SDH
+            elif disposition.get("forced"):
+                stype = SubtitleType.FORCED
+            else:
+                stype = SubtitleType.STANDARD
+
+            tracks.append(SubtitleTrack(
+                language=lang,
+                format=None,
+                subtitle_type=stype,
+                is_embedded=True,
+                raw_tokens=[lang_code] if lang_code else [],
+            ))
+
+        logger.debug(f"SubtitleIdentifier: {len(tracks)} embedded track(s) in {video_path.name}")
+        return tracks
+
+    # ------------------------------------------------------------------
+    # External tracks — filesystem scan per pattern strategy
+    # ------------------------------------------------------------------
+
+    def _scan_external(self, video_path: Path, pattern: SubtitlePattern) -> list[SubtitleTrack]:
+        strategy = pattern.scan_strategy
+
+        if strategy == ScanStrategy.ADJACENT:
+            candidates = self._find_adjacent(video_path)
+        elif strategy == ScanStrategy.FLAT:
+            candidates = self._find_flat(video_path, pattern.root_folder or "Subs")
+        elif strategy == ScanStrategy.EPISODE_SUBFOLDER:
+            candidates = self._find_episode_subfolder(video_path, pattern.root_folder or "Subs")
+        else:
+            return []
+
+        return self._classify_files(candidates, pattern)
+
+    def _find_adjacent(self, video_path: Path) -> list[Path]:
+        return [
+            p for p in sorted(video_path.parent.iterdir())
+            if p.is_file() and p.suffix.lower() in self.kb.known_extensions()
+            and p.stem != video_path.stem
+        ]
+
+    def _find_flat(self, video_path: Path, root_folder: str) -> list[Path]:
+        subs_dir = video_path.parent / root_folder
+        if not subs_dir.is_dir():
+            # Also look at release root (one level up)
+            subs_dir = video_path.parent.parent / root_folder
+        if not subs_dir.is_dir():
+            return []
+        return [
+            p for p in sorted(subs_dir.iterdir())
+            if p.is_file() and p.suffix.lower() in self.kb.known_extensions()
+        ]
+
+    def _find_episode_subfolder(self, video_path: Path, root_folder: str) -> list[Path]:
+        """
+        Look for Subs/{episode_stem}/*.srt
+
+        Checks two locations:
+          1. Adjacent to the video: video_path.parent / root_folder / video_path.stem
+          2. Release root (one level up): video_path.parent.parent / root_folder / video_path.stem
+        """
+        episode_stem = video_path.stem
+        candidates_dirs = [
+            video_path.parent / root_folder / episode_stem,
+            video_path.parent.parent / root_folder / episode_stem,
+        ]
+        for subs_dir in candidates_dirs:
+            if subs_dir.is_dir():
+                files = [
+                    p for p in sorted(subs_dir.iterdir())
+                    if p.is_file() and p.suffix.lower() in self.kb.known_extensions()
+                ]
+                if files:
+                    logger.debug(f"SubtitleIdentifier: found {len(files)} file(s) in {subs_dir}")
+                    return files
+        return []
+
+    # ------------------------------------------------------------------
+    # Classification
+    # ------------------------------------------------------------------
+
+    def _classify_files(self, paths: list[Path], pattern: SubtitlePattern) -> list[SubtitleTrack]:
+        tracks = []
+        for path in paths:
+            track = self._classify_single(path)
+            tracks.append(track)
+
+        # Post-process: if multiple tracks share same language but type is ambiguous,
+        # apply size_and_count disambiguation
+        if pattern.type_detection.value == "size_and_count":
+            tracks = self._disambiguate_by_size(tracks)
+
+        return tracks
+
+    def _classify_single(self, path: Path) -> SubtitleTrack:
+        fmt = self.kb.format_for_extension(path.suffix)
+        tokens = _tokenize(path.stem)
+
+        language = None
+        subtitle_type = SubtitleType.UNKNOWN
+        unknown_tokens = []
+        matched_tokens = 0
+
+        for token in tokens:
+            if self.kb.is_known_lang_token(token):
+                language = self.kb.language_for_token(token)
+                matched_tokens += 1
+            elif self.kb.is_known_type_token(token):
+                subtitle_type = self.kb.type_for_token(token) or subtitle_type
+                matched_tokens += 1
+            elif token.isdigit():
+                pass  # numeric prefix — ignore
+            elif len(token) > 1:
+                unknown_tokens.append(token)
+
+        # Confidence: proportion of meaningful tokens that were recognized
+        meaningful = [t for t in tokens if not t.isdigit() and len(t) > 1]
+        confidence = matched_tokens / max(len(meaningful), 1) if meaningful else 0.5
+
+        if unknown_tokens:
+            logger.debug(
+                f"SubtitleIdentifier: unknown tokens in '{path.name}': {unknown_tokens}"
+            )
+
+        size_kb = path.stat().st_size / 1024 if path.exists() else None
+        entry_count = _count_entries(path) if path.exists() else None
+
+        return SubtitleTrack(
+            language=language,
+            format=fmt,
+            subtitle_type=subtitle_type,
+            is_embedded=False,
+            file_path=path,
+            file_size_kb=size_kb,
+            entry_count=entry_count,
+            confidence=confidence,
+            raw_tokens=tokens,
+        )
+
+    def _disambiguate_by_size(self, tracks: list[SubtitleTrack]) -> list[SubtitleTrack]:
+        """
+        When multiple tracks share the same language and type is UNKNOWN/STANDARD,
+        the one with the most entries (lines) is SDH, the smallest is FORCED if
+        there are 3+, otherwise the smaller is STANDARD.
+
+        Only applied when type_detection = size_and_count.
+        """
+        from itertools import groupby
+
+        # Group by language code
+        lang_groups: dict[str, list[SubtitleTrack]] = {}
+        for track in tracks:
+            key = track.language.code if track.language else "__unknown__"
+            lang_groups.setdefault(key, []).append(track)
+
+        result = []
+        for lang_code, group in lang_groups.items():
+            if len(group) == 1:
+                result.extend(group)
+                continue
+
+            # Sort by entry_count ascending (None treated as 0)
+            sorted_group = sorted(group, key=lambda t: t.entry_count or 0)
+
+            if len(sorted_group) == 2:
+                # smaller = standard, larger = sdh
+                self._set_type(sorted_group[0], SubtitleType.STANDARD)
+                self._set_type(sorted_group[1], SubtitleType.SDH)
+            elif len(sorted_group) >= 3:
+                # smallest = forced, middle = standard, largest = sdh
+                self._set_type(sorted_group[0], SubtitleType.FORCED)
+                for t in sorted_group[1:-1]:
+                    self._set_type(t, SubtitleType.STANDARD)
+                self._set_type(sorted_group[-1], SubtitleType.SDH)
+
+            result.extend(sorted_group)
+
+        return result
+
+    def _set_type(self, track: SubtitleTrack, stype: SubtitleType) -> None:
+        """Mutate track type in-place."""
+        track.subtitle_type = stype

alfred/domain/subtitles/services/matcher.py

@@ -0,0 +1,118 @@
+"""SubtitleMatcher — filters tracks against resolved rules."""
+
+import logging
+
+from ..entities import SubtitleTrack
+from ..value_objects import SubtitleMatchingRules, SubtitleType
+
+logger = logging.getLogger(__name__)
+
+
+class SubtitleMatcher:
+    """
+    Filters a list of SubtitleTrack against effective SubtitleMatchingRules.
+
+    Returns matched tracks (pass all filters, confidence >= min_confidence)
+    and unresolved tracks (need user clarification).
+
+    Conflict resolution: when two tracks share the same language + type,
+    format_priority decides which one to keep.
+    """
+
+    def match(
+        self,
+        tracks: list[SubtitleTrack],
+        rules: SubtitleMatchingRules,
+    ) -> tuple[list[SubtitleTrack], list[SubtitleTrack]]:
+        """
+        Returns (matched, unresolved).
+        """
+        matched: list[SubtitleTrack] = []
+        unresolved: list[SubtitleTrack] = []
+
+        for track in tracks:
+            if track.is_embedded:
+                continue
+
+            if track.language is None or track.confidence < rules.min_confidence:
+                unresolved.append(track)
+                continue
+
+            if not self._passes_filters(track, rules):
+                logger.debug(f"SubtitleMatcher: filtered out {track}")
+                continue
+
+            matched.append(track)
+
+        matched = self._resolve_conflicts(matched, rules)
+        logger.info(
+            f"SubtitleMatcher: {len(matched)} matched, {len(unresolved)} unresolved"
+        )
+        return matched, unresolved
+
+    def _passes_filters(self, track: SubtitleTrack, rules: SubtitleMatchingRules) -> bool:
+        # Language filter
+        if rules.preferred_languages:
+            if not track.language:
+                return False
+            if track.language.code not in rules.preferred_languages:
+                return False
+
+        # Format filter (only for external files)
+        if rules.preferred_formats and not track.is_embedded:
+            if not track.format:
+                return False
+            if track.format.id not in rules.preferred_formats:
+                return False
+
+        # Type filter
+        if rules.allowed_types:
+            if track.subtitle_type.value not in rules.allowed_types:
+                return False
+
+        return True
+
+    def _resolve_conflicts(
+        self,
+        tracks: list[SubtitleTrack],
+        rules: SubtitleMatchingRules,
+    ) -> list[SubtitleTrack]:
+        """
+        When multiple tracks have same language + type, keep only the best one
+        according to format_priority. If no format_priority applies, keep the first.
+        """
+        seen: dict[tuple, SubtitleTrack] = {}
+
+        for track in tracks:
+            lang = track.language.code if track.language else None
+            stype = track.subtitle_type.value
+            key = (lang, stype)
+
+            if key not in seen:
+                seen[key] = track
+            else:
+                existing = seen[key]
+                if self._prefer(track, existing, rules.format_priority):
+                    logger.debug(
+                        f"SubtitleMatcher: conflict {key} — "
+                        f"preferring {track.format.id if track.format else 'embedded'} "
+                        f"over {existing.format.id if existing.format else 'embedded'}"
+                    )
+                    seen[key] = track
+
+        return list(seen.values())
+
+    def _prefer(
+        self,
+        candidate: SubtitleTrack,
+        existing: SubtitleTrack,
+        format_priority: list[str],
+    ) -> bool:
+        """Return True if candidate is preferable to existing."""
+        if not format_priority:
+            return False
+        c_fmt = candidate.format.id if candidate.format else ""
+        e_fmt = existing.format.id if existing.format else ""
+        c_rank = format_priority.index(c_fmt) if c_fmt in format_priority else 999
+        e_rank = format_priority.index(e_fmt) if e_fmt in format_priority else 999
+        return c_rank < e_rank

alfred/domain/subtitles/services/pattern_detector.py

@@ -0,0 +1,205 @@
+"""PatternDetector — discovers the subtitle structure of a release folder."""
+
+import json
+import logging
+import subprocess
+from pathlib import Path
+
+from ..knowledge.base import SubtitleKnowledgeBase
+from ..value_objects import ScanStrategy, SubtitlePattern
+
+logger = logging.getLogger(__name__)
+
+
+class PatternDetector:
+    """
+    Inspects a release folder and returns the best matching known pattern,
+    plus a confidence score and a description of what was found.
+
+    Used for "pattern discovery" — when we don't yet know which pattern
+    a release follows. The result is proposed to the user for confirmation.
+    """
+
+    def __init__(self, kb: SubtitleKnowledgeBase):
+        self.kb = kb
+
+    def detect(self, release_root: Path, sample_video: Path) -> dict:
+        """
+        Analyse the release folder and return:
+        {
+            "detected": SubtitlePattern | None,
+            "confidence": float,
+            "description": str,         # human-readable description of what was found
+            "candidate_pattern_ids": list[str],
+        }
+        """
+        findings = self._inspect(release_root, sample_video)
+        best, confidence = self._match_pattern(findings)
+
+        return {
+            "detected": best,
+            "confidence": confidence,
+            "description": self._describe(findings),
+            "candidate_pattern_ids": [best.id] if best else [],
+            "raw_findings": findings,
+        }
+
+    def _has_embedded_subtitles(self, video_path: Path) -> bool:
+        """Run ffprobe to check whether the video has embedded subtitle streams."""
+        try:
+            result = subprocess.run(
+                [
+                    "ffprobe", "-v", "quiet",
+                    "-print_format", "json",
+                    "-show_streams",
+                    "-select_streams", "s",
+                    str(video_path),
+                ],
+                capture_output=True, text=True, timeout=30,
+            )
+            data = json.loads(result.stdout)
+            return len(data.get("streams", [])) > 0
+        except (subprocess.TimeoutExpired, json.JSONDecodeError, FileNotFoundError):
+            return False
+
+    def _inspect(self, release_root: Path, sample_video: Path) -> dict:
+        """Gather structural facts about the release."""
+        known_exts = self.kb.known_extensions()
+        findings: dict = {
+            "has_subs_folder": False,
+            "subs_strategy": None,       # "flat" | "episode_subfolder"
+            "subs_root": None,
+            "adjacent_subs": False,
+            "has_embedded": self._has_embedded_subtitles(sample_video),
+            "files_per_episode": 0,
+            "has_lang_tokens": False,
+            "has_numeric_prefix": False,
+        }
+
+        # Check for Subs/ folder — adjacent or at release root
+        for subs_candidate in [
+            sample_video.parent / "Subs",
+            release_root / "Subs",
+        ]:
+            if subs_candidate.is_dir():
+                findings["has_subs_folder"] = True
+                findings["subs_root"] = str(subs_candidate)
+
+                # Is it flat or episode_subfolder?
+                children = list(subs_candidate.iterdir())
+                sub_files = [c for c in children if c.is_file() and c.suffix.lower() in known_exts]
+                sub_dirs = [c for c in children if c.is_dir()]
+
+                if sub_dirs and not sub_files:
+                    findings["subs_strategy"] = "episode_subfolder"
+                    # Count files in a sample subfolder
+                    sample_sub = sub_dirs[0]
+                    sample_files = [f for f in sample_sub.iterdir()
+                                    if f.is_file() and f.suffix.lower() in known_exts]
+                    findings["files_per_episode"] = len(sample_files)
+                    # Check naming conventions
+                    for f in sample_files:
+                        stem = f.stem
+                        parts = stem.split("_")
+                        if parts[0].isdigit():
+                            findings["has_numeric_prefix"] = True
+                        if any(self.kb.is_known_lang_token(t.lower())
+                               for t in stem.replace("_", ".").split(".")):
+                            findings["has_lang_tokens"] = True
+                else:
+                    findings["subs_strategy"] = "flat"
+                    findings["files_per_episode"] = len(sub_files)
+                    for f in sub_files:
+                        if any(self.kb.is_known_lang_token(t.lower())
+                               for t in f.stem.replace("_", ".").split(".")):
+                            findings["has_lang_tokens"] = True
+                break
+
+        # Check adjacent subs (next to the video)
+        if not findings["has_subs_folder"]:
+            adjacent = [
+                p for p in sample_video.parent.iterdir()
+                if p.is_file() and p.suffix.lower() in known_exts
+            ]
+            if adjacent:
+                findings["adjacent_subs"] = True
+                findings["files_per_episode"] = len(adjacent)
+
+        return findings
+
+    def _match_pattern(self, findings: dict) -> tuple[SubtitlePattern | None, float]:
+        """Score all known patterns against the findings."""
+        scores: list[tuple[float, SubtitlePattern]] = []
+
+        for pattern in self.kb.patterns().values():
+            score = self._score(pattern, findings)
+            scores.append((score, pattern))
+
+        if not scores:
+            return None, 0.0
+
+        scores.sort(key=lambda x: x[0], reverse=True)
+        best_score, best_pattern = scores[0]
+
+        if best_score < 0.4:
+            return None, best_score
+
+        return best_pattern, best_score
+
+    def _score(self, pattern: SubtitlePattern, findings: dict) -> float:
+        """Return a 0.0–1.0 match score for this pattern against the findings."""
+        score = 0.0
+        total = 0.0
+
+        strategy = pattern.scan_strategy
+
+        if strategy == ScanStrategy.EMBEDDED:
+            total += 1
+            if findings.get("has_embedded"):
+                score += 1.0
+            if not findings.get("has_subs_folder") and not findings.get("adjacent_subs"):
+                score += 0.5
+                total += 0.5
+
+        elif strategy == ScanStrategy.EPISODE_SUBFOLDER:
+            total += 3
+            if findings.get("has_subs_folder"):
+                score += 1.0
+            if findings.get("subs_strategy") == "episode_subfolder":
+                score += 2.0
+
+        elif strategy == ScanStrategy.FLAT:
+            total += 2
+            if findings.get("has_subs_folder"):
+                score += 1.0
+            if findings.get("subs_strategy") == "flat":
+                score += 1.0
+
+        elif strategy == ScanStrategy.ADJACENT:
+            total += 2
+            if findings.get("adjacent_subs"):
+                score += 1.0
+            if not findings.get("has_subs_folder"):
+                score += 1.0
+
+        return score / total if total > 0 else 0.0
+
+    def _describe(self, findings: dict) -> str:
+        parts = []
+        if findings.get("has_subs_folder"):
+            strategy = findings.get("subs_strategy", "?")
+            n = findings.get("files_per_episode", 0)
+            parts.append(f"Subs/ folder found ({strategy}), ~{n} file(s) per episode")
+            if findings.get("has_numeric_prefix"):
+                parts.append("files have numeric prefix (e.g. 2_English.srt)")
+            if findings.get("has_lang_tokens"):
+                parts.append("language tokens found in filenames")
+        elif findings.get("adjacent_subs"):
+            parts.append("subtitle files adjacent to video")
+        else:
+            parts.append("no external subtitle files found")
+
+        if findings.get("has_embedded"):
+            parts.append("embedded tracks detected (ffprobe)")
+
+        return " — ".join(parts) if parts else "nothing found"

alfred/domain/subtitles/services/placer.py

@@ -0,0 +1,93 @@
+"""SubtitlePlacer — hard-links matched subtitle tracks next to the destination video."""
+
+import logging
+import os
+from dataclasses import dataclass
+from pathlib import Path
+
+from ..entities import SubtitleTrack
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class PlacedTrack:
+    source: Path
+    destination: Path
+    filename: str
+
+
+@dataclass
+class PlaceResult:
+    placed: list[PlacedTrack]
+    skipped: list[tuple[SubtitleTrack, str]]   # (track, reason)
+
+    @property
+    def placed_count(self) -> int:
+        return len(self.placed)
+
+    @property
+    def skipped_count(self) -> int:
+        return len(self.skipped)
+
+
+class SubtitlePlacer:
+    """
+    Hard-links matched SubtitleTrack files next to a destination video.
+
+    Uses the same hard-link strategy as FileManager.copy_file:
+    instant, no data duplication, qBittorrent keeps seeding.
+
+    Embedded tracks are skipped — nothing to place on disk.
+    """
+
+    def place(
+        self,
+        tracks: list[SubtitleTrack],
+        destination_video: Path,
+    ) -> PlaceResult:
+        placed: list[PlacedTrack] = []
+        skipped: list[tuple[SubtitleTrack, str]] = []
+
+        dest_dir = destination_video.parent
+
+        for track in tracks:
+            if track.is_embedded:
+                logger.debug(f"SubtitlePlacer: skip embedded track ({track.language})")
+                skipped.append((track, "embedded — no file to place"))
+                continue
+
+            if not track.file_path or not track.file_path.exists():
+                skipped.append((track, "source file not found"))
+                continue
+
+            try:
+                dest_name = track.destination_name
+            except ValueError as e:
+                skipped.append((track, str(e)))
+                continue
+
+            dest_path = dest_dir / dest_name
+
+            if dest_path.exists():
+                logger.debug(f"SubtitlePlacer: skip {dest_name} — already exists")
+                skipped.append((track, "destination already exists"))
+                continue
+
+            try:
+                os.link(track.file_path, dest_path)
+                placed.append(PlacedTrack(
+                    source=track.file_path,
+                    destination=dest_path,
+                    filename=dest_name,
+                ))
+                logger.info(f"SubtitlePlacer: placed {dest_name}")
+            except OSError as e:
+                logger.warning(f"SubtitlePlacer: failed to place {dest_name}: {e}")
+                skipped.append((track, str(e)))
+
+        logger.info(
+            f"SubtitlePlacer: {len(placed)} placed, {len(skipped)} skipped "
+            f"for {destination_video.name}"
+        )
+        return PlaceResult(placed=placed, skipped=skipped)

alfred/domain/subtitles/services/utils.py

@@ -0,0 +1,21 @@
+"""Subtitle service utilities."""
+
+from ..entities import SubtitleTrack
+
+
+def available_subtitles(tracks: list[SubtitleTrack]) -> list[SubtitleTrack]:
+    """
+    Return the distinct subtitle tracks available, deduped by (language, type).
+
+    Useful to display what is available for a media item regardless of user
+    preferences — e.g. eng, eng.sdh, fra all show up as separate entries.
+    """
+    seen: set[tuple] = set()
+    result: list[SubtitleTrack] = []
+    for track in tracks:
+        lang = track.language.code if track.language else None
+        key = (lang, track.subtitle_type)
+        if key not in seen:
+            seen.add(key)
+            result.append(track)
+    return result

alfred/domain/subtitles/value_objects.py

@@ -0,0 +1,93 @@
+"""Subtitle domain value objects."""
+
+from dataclasses import dataclass, field
+from enum import Enum
+from pathlib import Path
+from typing import Any
+
+
+class ScanStrategy(Enum):
+    """How to locate subtitle files for a given release."""
+
+    ADJACENT = "adjacent"             # .srt next to the video
+    FLAT = "flat"                     # Subs/*.srt
+    EPISODE_SUBFOLDER = "episode_subfolder"  # Subs/{episode_name}/*.srt
+    EMBEDDED = "embedded"             # tracks inside the video container
+
+
+class TypeDetectionMethod(Enum):
+    """How to differentiate standard / SDH / forced when tokens are ambiguous."""
+
+    TOKEN_IN_NAME = "token_in_name"
+    SIZE_AND_COUNT = "size_and_count"
+    FFPROBE_METADATA = "ffprobe_metadata"
+
+
+class SubtitleType(Enum):
+    STANDARD = "standard"
+    SDH = "sdh"
+    FORCED = "forced"
+    UNKNOWN = "unknown"
+
+
+@dataclass(frozen=True)
+class SubtitleFormat:
+    """A known subtitle file format."""
+
+    id: str
+    extensions: list[str]
+    description: str = ""
+
+    def matches_extension(self, ext: str) -> bool:
+        return ext.lower() in [e.lower() for e in self.extensions]
+
+
+@dataclass(frozen=True)
+class SubtitleLanguage:
+    """A known subtitle language with its recognition tokens."""
+
+    code: str          # ISO 639-1
+    tokens: list[str]  # lowercase
+
+    def matches_token(self, token: str) -> bool:
+        return token.lower() in self.tokens
+
+
+@dataclass(frozen=True)
+class SubtitlePattern:
+    """
+    A known structural pattern for how a release group organises subtitle files.
+
+    Patterns are loaded from alfred/knowledge/patterns/*.yaml and are
+    independent of any specific release group — multiple groups can share
+    the same pattern.
+    """
+
+    id: str
+    description: str
+    scan_strategy: ScanStrategy
+    root_folder: str | None          # e.g. "Subs", None for adjacent/embedded
+    type_detection: TypeDetectionMethod
+    version: str = "1.0"
+
+
+@dataclass(frozen=True)
+class SubtitleMatchingRules:
+    """
+    Effective rules after scope resolution (global → group → show → season → episode).
+    Only stores actual values — None means "inherited, not overridden at this level".
+    """
+
+    preferred_languages: list[str] = field(default_factory=list)   # ISO 639-1 codes
+    preferred_formats: list[str] = field(default_factory=list)     # format ids
+    allowed_types: list[str] = field(default_factory=list)         # SubtitleType ids
+    format_priority: list[str] = field(default_factory=list)       # ordered format ids
+    min_confidence: float = 0.7
+
+
+@dataclass(frozen=True)
+class RuleScope:
+    """At which level a rule set applies."""
+
+    level: str   # "global" | "release_group" | "movie" | "show" | "season" | "episode"
+    identifier: str | None = None   # imdb_id, group name, "S01", "S01E03"…

alfred/domain/tv_shows/entities.py

@@ -2,7 +2,6 @@
 
 import re
 from dataclasses import dataclass, field
-from datetime import datetime
 
 from ..shared.value_objects import FilePath, FileSize, ImdbId
 from .value_objects import EpisodeNumber, SeasonNumber, ShowStatus
@@ -22,8 +21,6 @@ class TVShow:
     seasons_count: int
     status: ShowStatus
     tmdb_id: int | None = None
-    first_air_date: str | None = None
-    added_at: datetime = field(default_factory=datetime.now)
 
     def __post_init__(self):
         """Validate TV show entity."""
@@ -87,9 +84,6 @@ class Season:
     season_number: SeasonNumber
     episode_count: int
     name: str | None = None
-    overview: str | None = None
-    air_date: str | None = None
-    poster_path: str | None = None
 
     def __post_init__(self):
         """Validate season entity."""
@@ -146,11 +140,6 @@ class Episode:
     title: str
     file_path: FilePath | None = None
     file_size: FileSize | None = None
-    overview: str | None = None
-    air_date: str | None = None
-    still_path: str | None = None
-    vote_average: float | None = None
-    runtime: int | None = None  # in minutes
 
     def __post_init__(self):
         """Validate episode entity."""

alfred/infrastructure/api/knaben/client.py

@@ -6,7 +6,7 @@ from typing import Any
 import requests
 from requests.exceptions import HTTPError, RequestException, Timeout
 
-from agent.config import Settings, settings
+from alfred.settings import Settings, settings
 
 from .dto import TorrentResult
 from .exceptions import KnabenAPIError, KnabenNotFoundError

alfred/infrastructure/api/qbittorrent/client.py

@@ -6,7 +6,7 @@ from typing import Any
 import requests
 from requests.exceptions import HTTPError, RequestException, Timeout
 
-from agent.config import Settings, settings
+from alfred.settings import Settings, settings
 
 from .dto import TorrentInfo
 from .exceptions import QBittorrentAPIError, QBittorrentAuthError

alfred/infrastructure/api/tmdb/client.py

@@ -6,7 +6,7 @@ from typing import Any
 import requests
 from requests.exceptions import HTTPError, RequestException, Timeout
 
-from agent.config import Settings, settings
+from alfred.settings import Settings, settings
 
 from .dto import MediaResult
 from .exceptions import (

alfred/infrastructure/filesystem/file_manager.py

@@ -0,0 +1,317 @@
+"""File manager for filesystem operations."""
+
+import logging
+import os
+from collections import namedtuple
+from pathlib import Path
+from typing import Any
+
+from alfred.infrastructure.persistence import get_memory
+
+from .exceptions import PathTraversalError
+
+logger = logging.getLogger(__name__)
+
+FileOperationResult = namedtuple("FileOperationResult", ["success", "error", "message"])
+
+
+def _err(error: str, message: str) -> dict[str, Any]:
+    return {"status": "error", "error": error, "message": message}
+
+
+class FileManager:
+    """
+    File manager for filesystem operations.
+
+    Handles folder configuration, listing, and file operations
+    with security checks to prevent path traversal attacks.
+    """
+
+    def set_folder_path(self, folder_name: str, path_value: str) -> dict[str, Any]:
+        """
+        Set a folder path in the configuration.
+
+        Args:
+            folder_name: Name of folder (download, tvshow, movie, torrent).
+            path_value: Absolute path to the folder.
+
+        Returns:
+            Dict with status or error information.
+        """
+        try:
+            path_obj = Path(path_value).resolve()
+
+            if not path_obj.exists():
+                return _err("invalid_path", f"Path does not exist: {path_value}")
+
+            if not path_obj.is_dir():
+                return _err("invalid_path", f"Path is not a directory: {path_value}")
+
+            if not os.access(path_obj, os.R_OK):
+                return _err("permission_denied", f"Path is not readable: {path_value}")
+
+            memory = get_memory()
+            # workspace folders have fixed attributes; library folders go in the dict
+            if folder_name in ("download", "torrent"):
+                setattr(memory.ltm.workspace, folder_name, str(path_obj))
+            else:
+                memory.ltm.library_paths.set(folder_name, str(path_obj))
+            memory.save()
+
+            logger.info(f"Set {folder_name} to: {path_obj}")
+            return {"status": "ok", "folder_name": folder_name, "path": str(path_obj)}
+
+        except ValueError as e:
+            return _err("validation_failed", str(e))
+
+        except Exception as e:
+            logger.error(f"Unexpected error setting path: {e}", exc_info=True)
+            return _err("internal_error", "Failed to set path")
+
+    def list_folder(self, folder_type: str, path: str = ".") -> dict[str, Any]:
+        """
+        List contents of a configured folder.
+
+        Args:
+            folder_type: Type of folder (download, tvshow, movie, torrent).
+            path: Relative path within the folder (default: root).
+
+        Returns:
+            Dict with folder contents or error information.
+        """
+        try:
+            safe_path = self._sanitize_path(path)
+
+            memory = get_memory()
+            if folder_type in ("download", "torrent"):
+                folder_path = getattr(memory.ltm.workspace, folder_type, None)
+            else:
+                folder_path = memory.ltm.library_paths.get(folder_type)
+
+            if not folder_path:
+                return _err("folder_not_set", f"{folder_type.capitalize()} folder not configured.")
+
+            root = Path(folder_path)
+            target = root / safe_path
+
+            if not self._is_safe_path(root, target):
+                return _err("forbidden", "Access denied: path outside allowed directory")
+
+            if not target.exists():
+                return _err("not_found", f"Path does not exist: {safe_path}")
+
+            if not target.is_dir():
+                return _err("not_a_directory", f"Path is not a directory: {safe_path}")
+
+            try:
+                entries = [entry.name for entry in target.iterdir()]
+                logger.debug(f"Listed {len(entries)} entries in {target}")
+                return {
+                    "status": "ok",
+                    "folder_type": folder_type,
+                    "path": safe_path,
+                    "entries": sorted(entries),
+                    "count": len(entries),
+                }
+            except PermissionError:
+                return _err("permission_denied", f"Permission denied: {safe_path}")
+
+        except PathTraversalError as e:
+            return _err("forbidden", str(e))
+
+        except ValueError as e:
+            return _err("validation_failed", str(e))
+
+        except Exception as e:
+            logger.error(f"Unexpected error listing folder: {e}", exc_info=True)
+            return _err("internal_error", "Failed to list folder")
+
+    def copy_file(self, source: str, destination: str) -> dict[str, Any]:
+        """
+        Hard-link a file to a destination (instant, no data duplication).
+
+        Both paths must be on the same filesystem. qBittorrent keeps seeding
+        the original inode unaffected.
+
+        Args:
+            source: Absolute path to the source file.
+            destination: Absolute path to the destination file.
+
+        Returns:
+            Dict with status or error information.
+        """
+        try:
+            source_path = Path(source).resolve()
+            dest_path = Path(destination).resolve()
+
+            logger.info(f"Hard-linking: {source_path} -> {dest_path}")
+
+            if not source_path.exists():
+                return _err("source_not_found", f"Source does not exist: {source}")
+
+            if not source_path.is_file():
+                return _err("source_not_file", f"Source is not a file: {source}")
+
+            if not dest_path.parent.exists():
+                return _err("destination_dir_not_found", f"Destination directory does not exist: {dest_path.parent}")
+
+            if dest_path.exists():
+                return _err("destination_exists", f"Destination already exists: {destination}")
+
+            os.link(source_path, dest_path)
+
+            logger.info(f"Hard link created: {dest_path.name}")
+            return {
+                "status": "ok",
+                "source": str(source_path),
+                "destination": str(dest_path),
+                "filename": dest_path.name,
+                "size": source_path.stat().st_size,
+            }
+
+        except OSError as e:
+            logger.error(f"Error creating hard link: {e}", exc_info=True)
+            return _err("link_failed", str(e))
+
+    def move_file(self, source: str, destination: str) -> dict[str, Any]:
+        """
+        Move a file via hard link + source deletion.
+
+        Hard-links the file to the destination, then removes the source.
+        qBittorrent keeps seeding during the operation since the inode
+        is still referenced until the source is removed.
+
+        Args:
+            source: Absolute path to the source file.
+            destination: Absolute path to the destination file.
+
+        Returns:
+            Dict with status or error information.
+        """
+        try:
+            source_path = Path(source).resolve()
+
+            link_result = self.copy_file(source, destination)
+            if link_result.get("status") != "ok":
+                return link_result
+
+            source_path.unlink()
+
+            logger.info(f"File moved: {source_path.name} -> {link_result['destination']}")
+            return {
+                "status": "ok",
+                "source": str(source_path),
+                "destination": link_result["destination"],
+                "filename": link_result["filename"],
+                "size": link_result["size"],
+            }
+
+        except Exception as e:
+            logger.error(f"Error moving file: {e}", exc_info=True)
+            return _err("move_failed", str(e))
+
+    def create_seed_links(
+        self, library_file: str, original_download_folder: str, torrent_folder: str
+    ) -> dict[str, Any]:
+        """
+        Prepare a torrent folder so qBittorrent can keep seeding after a move.
+
+        - Hard-links the moved video file from the library back into
+          torrents/<original_folder_name>/ (same inode, no data copy).
+        - Copies every other file from the original download folder
+          (.srt, .nfo, .jpg, .txt, …) into the same torrent subfolder,
+          preserving relative paths.
+
+        Args:
+            library_file: Absolute path to the video file in the library.
+            original_download_folder: Absolute path to the download folder
+                that contained the original release (may still have subs etc.).
+            torrent_folder: Absolute path to the root torrents/ directory.
+
+        Returns:
+            Dict with status, linked_file, copied_files list, skipped list.
+        """
+        try:
+            lib_path = Path(library_file).resolve()
+            src_folder = Path(original_download_folder).resolve()
+            torrent_root = Path(torrent_folder).resolve()
+
+            if not lib_path.exists():
+                return _err("library_file_not_found", f"Library file not found: {library_file}")
+            if not src_folder.exists():
+                return _err("source_folder_not_found", f"Download folder not found: {original_download_folder}")
+            if not torrent_root.exists():
+                return _err("torrent_folder_not_found", f"Torrent folder not found: {torrent_folder}")
+
+            dest_folder = torrent_root / src_folder.name
+            dest_folder.mkdir(parents=True, exist_ok=True)
+
+            # Hard-link the video file from library → torrent subfolder
+            link_dest = dest_folder / lib_path.name
+            if link_dest.exists():
+                return _err("destination_exists", f"Link already exists: {link_dest}")
+            os.link(lib_path, link_dest)
+            logger.info(f"Hard-linked for seeding: {lib_path.name} → {dest_folder}")
+
+            # Copy everything else from the original download folder
+            copied: list[str] = []
+            skipped: list[str] = []
+            for item in src_folder.rglob("*"):
+                if not item.is_file():
+                    continue
+                rel = item.relative_to(src_folder)
+                dest_item = dest_folder / rel
+                dest_item.parent.mkdir(parents=True, exist_ok=True)
+                if dest_item.exists():
+                    skipped.append(str(rel))
+                    continue
+                import shutil
+                shutil.copy2(item, dest_item)
+                copied.append(str(rel))
+                logger.debug(f"Copied for seeding: {rel}")
+
+            return {
+                "status": "ok",
+                "torrent_subfolder": str(dest_folder),
+                "linked_file": str(link_dest),
+                "copied_files": copied,
+                "copied_count": len(copied),
+                "skipped": skipped,
+            }
+
+        except OSError as e:
+            logger.error(f"create_seed_links failed: {e}", exc_info=True)
+            return _err("link_failed", str(e))
+        except Exception as e:
+            logger.error(f"create_seed_links unexpected error: {e}", exc_info=True)
+            return _err("internal_error", str(e))
+
+    def _sanitize_path(self, path: str) -> str:
+        """
+        Sanitize a relative path to prevent path traversal attacks.
+
+        Raises PathTraversalError if the path tries to escape the root.
+        """
+        normalized = os.path.normpath(path)
+
+        # Reject absolute paths
+        if os.path.isabs(normalized):
+            raise PathTraversalError("Absolute paths are not allowed")
+
+        # Reject parent directory references
+        if normalized.startswith("..") or "/.." in normalized or "\\.." in normalized:
+            raise PathTraversalError("Parent directory references not allowed")
+
+        # Reject null bytes
+        if "\x00" in normalized:
+            raise PathTraversalError("Null bytes in path not allowed")
+
+        return normalized
+
+    def _is_safe_path(self, base_path: Path, target_path: Path) -> bool:
+        """Return True if target_path is inside base_path (prevents traversal)."""
+        try:
+            target_path.resolve().relative_to(base_path.resolve())
+            return True
+        except (ValueError, OSError):
+            return False

alfred/infrastructure/filesystem/organizer.py

@@ -3,9 +3,9 @@
 import logging
 from pathlib import Path
 
-from domain.movies.entities import Movie
+from alfred.domain.movies.entities import Movie
-from domain.tv_shows.entities import Episode, Season, TVShow
+from alfred.domain.tv_shows.entities import Episode, Season, TVShow
-from domain.tv_shows.value_objects import SeasonNumber
+from alfred.domain.tv_shows.value_objects import SeasonNumber
 
 logger = logging.getLogger(__name__)