Engram
Persistent, verbatim, searchable memory for AI assistants — one memory across every MCP client.
- Transport
- Not stated
- Package
- —
- Registry id
- app.getengram/engram
No install snippet on purpose. A working MCP config is a command, its arguments and an environment block — the last two are where API keys live, so this catalogue never stores them and cannot publish them. Follow the link above for the authors' own instructions.
It stores information word-for-word without summarization, auto-redacts secrets and PII, and makes the data searchable across different AI assistants. Engram is built on the Model Context Protocol and can connect any MCP-compatible agent in minutes. It uses an edge-first architecture with Cloudflare Workers, D1, and Vectorize for sub-50ms global latency and automatic scaling to zero when idle.
The system automatically strips API keys, tokens, and PII from saved memory to prevent credential leaks. Engram offers a free tier with 10,000 messages of permanent memory, unlimited conversations, and semantic search.
Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.