# LogicMem by LogicFrame

> Stop re-explaining your project to every AI. Say it once and every agent remembers: one shared memory for all your agents, connected in 5 minutes, with fewer mistakes (contradictions and gaps get flagged) and a tamper-evident receipt for every memory. One live memory shared by agents from different vendors (Claude, GPT, Cursor, OpenClaw, local models) over MCP, backed by a tamper-evident, hash-chained audit log (about 800k entries verify in our own production chain).

Key facts:
- MCP endpoint: https://mcp.logicmem.io/mcp (streamable HTTP, Bearer API key)
- Contradiction flagging, gap detection, usefulness-weighted recall
- Can be self-hosted with a local Qwen model
- Open-source client kit (MIT): https://github.com/LogicMem/LogicMem-mcp-
- Stage: early; recruiting design partners. No public pricing yet.

## Pages
- [Quickstart](https://logicmem.io/quickstart/): connect your whole agent team to one memory
- [LogicMem vs Mem0](https://logicmem.io/compare/mem0/)
- [LogicMem vs Cognee](https://logicmem.io/compare/cognee/)
- [LogicMem vs built-in ChatGPT/Claude memory](https://logicmem.io/compare/built-in-memory/)
- [EU AI Act Article 12 explainer](https://logicmem.io/eu-ai-act-article-12/)
- [Blog](https://logicmem.io/blog/)
- [Become a design partner](https://logicmem.io/design-partner/)