Description

Adds an Open Source tab to the Email Automation cookbook (docs/cookbooks/operations/email-automation.mdx), giving users a fully local alternative alongside the existing Platform version.

The OSS version uses:

• Mem0 Memory class with Memory.from_config() instead of MemoryClient
• Qdrant (local Docker) as the vector store
• Ollama (llama3.1:latest) for LLM and (nomic-embed-text:latest) for embeddings

Key API differences handled:

• MemoryClient() → Memory.from_config({...}) with Qdrant + Ollama providers
• categories=["email", "correspondence"] (Platform-only) → metadata={"memory_category": "email"} for OSS
• client.search(query, filters={"AND": [{"user_id": ...}, {"categories": ...}]}) → memory.search(query, user_id=..., filters={"memory_category": "email"}) (explicit session params, metadata-based
  category filtering)
• client.get_all(filters={"AND": [{"user_id": ...}, {"subject": {"icontains": ...}}]}) → memory.get_all(user_id=..., filters={"AND": [{"memory_category": "email"}, {"subject": {"icontains":
  ...}}]}) (icontains works natively in OSS via Qdrant's MatchText)

Includes installation section with pip install, Docker, and ollama pull commands, plus a note about swappable embedding models and a callout explaining the categories vs metadata difference.

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Refactor (no functional changes)
• Documentation update

Breaking Changes

N/A

Test Coverage

• I added/updated unit tests
• I added/updated integration tests
• I tested manually (describe below)
• No tests needed (explain why)

Tested the OSS cookbook code end-to-end locally with Ollama + Qdrant (5 tests):

1. memory.add() — stored Alice's email, extracted 3 memories (meeting day, time, presentation task)
2. memory.add() — stored Charlie's email, Mem0 intelligently deduped/updated existing memories
3. memory.search() with {"memory_category": "email"} filter — found 2 relevant results
4. memory.search() with compound AND filter (memory_category + sender) — correctly returned only Alice's email
5. memory.get_all() with {"subject": {"icontains": "Meeting"}} — found the meeting thread

Checklist

• My code follows the project's style guidelines
• I have performed a self-review of my code
• I have added tests that prove my fix/feature works
• New and existing tests pass locally
• I have updated documentation if needed