#377377th of 455 in Skills, plugins and prompts

mem9

Shared persistent memory server and plugins for coding agents

Stars
1.2k
License
Apache-2.0
Last commit
Nov 2026
Language
TypeScript
Stars, last 29 weeks+256
21 Apr 2026: 982 stars28 Apr 2026: 1,010 stars5 May 2026: 1,039 stars12 May 2026: 1,058 stars19 May 2026: 1,073 stars26 May 2026: 1,091 stars2 Jun 2026: 1,098 stars9 Jun 2026: 1,109 stars16 Jun 2026: 1,118 stars23 Jun 2026: 1,130 stars30 Jun 2026: 1,143 stars7 Jul 2026: 1,149 stars14 Jul 2026: 1,151 stars21 Jul 2026: 1,160 stars28 Jul 2026: 1,160 stars4 Aug 2026: 1,164 stars11 Aug 2026: 1,174 stars18 Aug 2026: 1,179 stars25 Aug 2026: 1,193 stars1 Sept 2026: 1,196 stars8 Sept 2026: 1,202 stars15 Sept 2026: 1,210 stars22 Sept 2026: 1,211 stars29 Sept 2026: 1,217 stars6 Oct 2026: 1,219 stars13 Oct 2026: 1,221 stars20 Oct 2026: 1,225 stars27 Oct 2026: 1,229 stars3 Nov 2026: 1,236 stars10 Nov 2026: 1,238 stars

982 on 21 Apr 20261,238 on 10 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

mem9 is a Go server (mnemo-server) that stores, searches, updates and deletes agent memories over a REST API, backed by TiDB, PostgreSQL or db9. Plugins connect OpenClaw, Hermes Agent, Claude Code, OpenCode, Codex, DeepSeek Harness and Dify to the same memory space. Search combines vector and full-text recall, and an optional LLM step extracts facts from conversations at ingest.

Who it is for: Engineers adding shared long-term memory to coding agents and workflows

Strengths

  • One memory space shared across several agent runtimes via a common REST API
  • Self-host on TiDB, PostgreSQL or db9, or use the hosted API at api.mem9.ai
  • Smart ingest extracts facts through any OpenAI-compatible LLM endpoint
  • Webhooks and Space Chains support multi-space recall and routing

Weaknesses

  • Postgres and db9 backends need manual tenant bootstrap; auto-provisioning covers TiDB only
  • Built-in auto-embedding is tied to TiDB Cloud Starter on AWS
  • No tagged release, and the API is still versioned v1alpha1/v1alpha2
  • Self-hosting requires applying a control-plane schema by hand before first start

What it needs

  • no GPU
  • Needs TiDB, PostgreSQL or db9, OpenAI-compatible LLM endpoint (optional, for smart ingest), Embedding provider (optional, or TiDB auto-embedding)
  • Models: OpenAI-compatible chat endpoints (default gpt-4o-mini), text-embedding-3-small (default embeddings), Cohere embed-multilingual-v3 via TiDB auto-embedding
  • port 8080

Also in Skills, plugins and prompts

See all 455
Also in Skills, plugins and prompts
RankProjectScore
1claude-memPersistent cross-session memory plugin for Claude Code and other coding agents116.4k stars+16,587 (+17%) in 30 daysApache-2.097 out of 100
2impeccableDesign skill with 24 slash commands for AI coding agents' frontend output Live demo ↗ (opens in a new tab)108.8k stars+28,464 (+35%) in 30 daysApache-2.097 out of 100
3gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles141.3k stars+5,544 (+4.1%) in 30 daysMIT96 out of 100
#4rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex Live demo ↗ (opens in a new tab)81.5k stars+7,374 (+9.9%) in 30 daysMIT96 out of 100
#5cavemanShortens coding-agent replies and compresses tool output before the model reads it Live demo ↗ (opens in a new tab)119.3k stars+8,422 (+7.6%) in 30 daysApache-2.095 out of 100