by NickCirv · MCP Server · ★ 134
Engram makes your AI coding agent stop re-reading the same files — and stop repeating mistakes your repo already fixed. One install. 8 IDEs. Up to 89% fewer context tokens vs reading whole files — typically 50–90%, strongest on large-file codebases (Python/Go), lower on many-small-file repos. Structural packet size, not a bill saving; run engram bench on yours. Bi-temporal mistakes auto-captured from git revert history. Local SQLite, zero cloud, Apache 2.0.
| Stars | 134 |
| Forks | 13 |
| Language | TypeScript |
| Category | MCP Server |
| License | Apache-2.0 |
| Quality Score | 63.7048320886665/100 |
| Open Issues | 11 |
| Last Updated | 2026-06-22 |
| Created | 2026-04-09 |
| Platforms | claude-code, cli, codex, mcp, node |
| Est. Tokens | ~25k |
Explore other popular mcp server tools:
engram is The context spine that 10x's every AI coding session. Live in 8 IDEs (Claude Code, Cursor, Cline, Continue, Aider, Codex, Windsurf, Zed) via npm + OpenVSX + Anthropic plugin directory. 89% measured to. It is categorized as a MCP Server with 134 GitHub stars.
engram is primarily written in TypeScript. It covers topics such as agent-memory, agent-tools, agents-md.
You can find installation instructions and usage details in the engram GitHub repository at github.com/NickCirv/engram. The project has 134 stars and 13 forks, indicating an active community.
engram is released under the Apache-2.0 license, making it free to use and modify according to the license terms.