Skip to content

Memnex documentation

Pick the doc that matches what you're trying to do.

By role

If you are... Start with
Integrating an AI agent with Memnex (most common) integration-guide.md
Running Memnex itself (self-hosting or as service operator) operator-guide.md
Contributing to Memnex (extending, fixing, releasing) ../CONTRIBUTING.md + publishing.md
Just exploring The project README, then architecture.md

Reference docs

Doc What's in it
architecture.md System design, data flow, the four invariants. Read this to understand how Memnex works internally.
mcp.md Full MCP API reference — 5 tools, 3 resources, 3 prompts, schemas, error shapes.
api-reference.md REST API, Python embedded API, CLI commands. Audience-tagged so each section is clearly tenant- or operator-facing.
security.md Multi-tenant isolation, audit ledger, regulated PII handling, injection floor.
features.md What's built today and what's intentionally out of scope.
deployment.md Docker compose, scaling, env vars.
eval.md The six benchmark suites and how to run them.
comparison.md How Memnex compares to Mem0, Zep, Letta, OpenAI memory.
roadmap.md What's next and why, in three priority tiers.
publishing.md How releases get to PyPI. Maintainer-facing.

Suggested reading paths

First-time tenant integrating an agent: integration-guide.mdmcp.md → done.

Operator setting up self-hosted Memnex: operator-guide.mddeployment.mdsecurity.md.

Engineer evaluating Memnex against alternatives: architecture.mdcomparison.mdfeatures.mdroadmap.md.

Maintainer cutting a release: ../CONTRIBUTING.mdpublishing.md../CHANGELOG.md.