Enterprise DNA
M MCP Servers Developer low

20alexl/claude-engram

by Various

Persistent memory and session intelligence for AI coding assistants. Auto-tracks mistakes, decisions, and context via hooks. Mines your full session history for patterns, predictio

2

MCP

20alexl/claude-engram

Added 1 June 2026

#claude #claude-code #developer-tool #hooks #mcp-server #model-control-protocol #obsisidan #ollama

Overview

Claude-Engram is a Python tool that provides persistent memory for AI coding assistants by hooking into sessions to automatically track mistakes, decisions, and context. It mines full session history to surface patterns, predictions, and enable cross-session search.

Best for

Best for
Developers who want persistent context and history search for AI coding assistants

Use cases

  • Track recurring coding errors across multiple AI-assisted sessions
  • Recall past design decisions without repeating them
  • Search through previous assistant interactions for relevant context

Notes

Claude-Engram is a Python tool that provides persistent memory for AI coding assistants by hooking into sessions to automatically track mistakes, decisions, and context. It mines full session history to surface patterns, predictions, and enable cross-session search.

15 stars on GitHub. Last updated 2026-05-29. Licensed MIT.

Use cases

  • Track recurring coding errors across multiple AI-assisted sessions
  • Recall past design decisions without repeating them
  • Search through previous assistant interactions for relevant context

Pros

  • Automatically captures session context without manual logging
  • Supports cross-session search, making history reusable
  • Lightweight hook-based integration with existing workflows

Cons

  • Small community and limited support due to only 15 stars
  • Requires manual set up of hooks into coding assistant sessions
  • Scope is limited to coding assistant interactions, not general memory

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.

Pros

  • Automatically captures session context without manual logging
  • Supports cross-session search, making history reusable
  • Lightweight hook-based integration with existing workflows

Cons

  • Small community and limited support due to only 15 stars
  • Requires manual set up of hooks into coding assistant sessions
  • Scope is limited to coding assistant interactions, not general memory