Bumblebiber/hmem

Humanlike persistent memory for AI agents — MCP server with 5-level lazy-loaded SQLite memory. Works with Claude Code, Gemini CLI, Cursor, Windsurf, OpenCode.

25
/ 100
Experimental

This project gives your AI agent a persistent, 'humanlike' memory that spans across different sessions, devices, and AI tools. It automatically records conversations and project details, allowing your agent to recall past decisions, errors, and progress without needing to be re-briefed. This is for developers, operations engineers, or anyone who uses AI agents for coding or complex tasks and wants their AI to maintain context over time.

Use this if your AI agent repeatedly forgets important context between sessions or across different development environments, leading to redundant work and wasted tokens.

Not ideal if you only use AI for one-off queries or simple tasks that don't require long-term memory or cross-project context.

AI-agent-development software-development developer-tools code-generation AI-assisted-coding
No Package No Dependents
Maintenance 10 / 25
Adoption 4 / 25
Maturity 11 / 25
Community 0 / 25

How are scores calculated?

Stars

7

Forks

Language

TypeScript

License

MIT

Last pushed

Mar 06, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mcp/Bumblebiber/hmem"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.