#137137th of 149 in Skills, plugins and rules
longmemory
Temporal, access-controlled memory store for LLM agents, backed by SQLite
- Stars
- 4.5k
- License
- Apache-2.0
- Last commit
- Sep 2026
- Last release
- Dec 2025
- Language
- TypeScript
Missed the latest GitHub check; these facts may be older.
Overview
LongMemory is a TypeScript memory engine that ingests events, documents and coding-agent sessions into an immutable graph stored in SQLite. It offers strict, historical, associative and world-grounded recall with separate recorded and valid time. The engine is exposed as an npm library, CLI, HTTP API, MCP server, Next.js dashboard and VS Code extension, plus a zero-dependency Python HTTP client.
Who it is for: Developers adding persistent memory to coding agents or LLM apps
Strengths
- Recorded time and valid time are separate, so historical recall returns superseded facts
- Runs in-memory or on SQLite with no external database; reopening restores state
- One engine across npm, CLI, HTTP, MCP (13 tools), dashboard and VS Code
- Imports sessions from Claude Code, Codex, OpenCode, Gemini CLI, Copilot Chat and Cline
Weaknesses
- Python package is only an HTTP client; the engine needs the TypeScript service
- Last tagged release is 2025-12-12, well behind recent commits
- Embedding providers must be configured separately (OpenAI-compatible, Gemini, Bedrock, Ollama)
- Hosted deployments require setting an API key and mounting persistent storage at /data
What it needs
- no GPU
- Docker + Compose
- Needs SQLite, Node.js, pnpm
- Models: OpenAI-compatible APIs, Gemini, AWS Bedrock, Ollama, Siray
- port 7331
Also in Skills, plugins and rules
See all 149| Rank | Project | Score |
|---|---|---|
| 1 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 95 out of 100 |
| 2 | skillsAgent skills for planning, TDD, debugging and code review | 94 out of 100 |
| 3 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 94 out of 100 |
| #4 | impeccableDesign skill with 24 commands and a detector for AI-generated frontends | 94 out of 100 |
| #5 | ponytailSkill prompt that makes coding agents write less code and add tests | 92 out of 100 |