Claude-Mem thedotmack/claude-mem

Memory plugin for Claude Code and other coding agents that captures tool use, summarises sessions and injects context into new sessions.

Creator kit + Compare GitHub ↗ Install guide (official) ↗

Claude-Mem at a glance

Checked 15 Oct 2026

Claude-Mem hooks into Claude Code (and OpenCode, Codex via T3 Code, Pi and others) to record observations from each session into a local SQLite database with Chroma vector search, then feeds summaries back into later sessions and exposes MCP search tools. Code is Apache-2.0 (open core; hosted sync and team features are reserved). Summaries are written by an LLM: your Anthropic plan, an OpenRouter or Gemini key, or the paid claude-mem observer after a 30-day trial. Needs Node.js 20+. It is coding-agent memory only, not a general memory API.

Cite this pageRepoQuarry. “Claude-Mem.” Checked 15 Oct 2026. https://repoquarry.dev/projects/claude-mem

How much of Mem0 Platform does it replace?

1 of 3 jobs. Memory for coding agents only; still needs a paid LLM plan or API key to write summaries.

Long-term memory for chat assistants Not covered

Memory for coding agents Partial replacement

Knowledge graph memory Not covered

For long-term memory for chat assistants, knowledge graph memory, see other Mem0 Platform alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction

QuestionAnswerSourceChecked

Runs on my machine? Yes, local worker service with a SQLite database Local HTTP API with web viewer UI and search endpoints, managed by Bun github.com · claude-mem#how-it-works ↗ 1d ago

Code licence Apache-2.0 (open core; hosted cloud and enterprise features are not in the repo) Claude-Mem uses an open-core structure. github.com · ip-boundary.md ↗ 1d ago

Model weights licence No bundled model; uses the LLM provider you pick github.com · claude-mem#quick-start ↗ 1d ago

Commercial use Yes, Apache-2.0 code github.com · LICENSE ↗ 1d ago

Paid services needed? Needs a coding agent and an LLM (Anthropic plan, OpenRouter or Gemini key, or paid claude-mem observer); no local model option documented After sign-in you pick your memory provider — the claude-mem observer, your own OpenRouter or Gemini key, or your Anthropic plan. github.com · claude-mem#quick-start ↗ 1d ago

Hardware Not documented; software needs Node.js 20+, Bun and uv github.com · claude-mem#system-requirements ↗ 1d ago

Operating systems Windows setup notes in the README; macOS and Linux not stated explicitly Claude Code hooks do not need Git Bash. github.com · claude-mem#windows-setup-notes ↗ 1d ago

Setup effort Easy, npx claude-mem install or the Claude Code plugin marketplace npx claude-mem install github.com · claude-mem#quick-start ↗ 1d ago

What it costs to run

Software$0

Required paid servicesSupported coding agent such as Claude Code, LLM provider (Anthropic plan, OpenRouter or Gemini key)

Optional paid servicesclaude-mem observer subscription, cmem.ai cloud sync

Your hardwareNot documented; software needs Node.js 20+, Bun and uv

Mem0 Platform: Free Hobby tier (10,000 add and 1,000 retrieval requests a month); Starter $19/month, Pro $249/month, Enterprise custom (source, checked 15 Oct 2026). Running it yourself costs your hardware, electricity and setup time.

Project activity

101kstars

8.9kforks

281contributors

Last code changetoday

Latest releasev13.39.0 · 3d ago

ArchivedNo

From GitHub, today. Stars show popularity, not quality.

Using an AI agent?

The same record, as JSON or through MCP.

get_project("claude-mem")
GET /api/v1/projects/claude-mem.json

API and MCP docs →