#161 · Primary category: AI Coding Assistants
Graft
Turbocharge Claude Code, Cursor, Codex, Gemini & every coding agent: faster, cheaper, with contextual understanding specific to your codebase.
Project last updated:08/29/26
GitHub Stars
5.1K
Forks
462
Contributors
30
License
MIT
Why we included this project
Most coding agents rediscover your repository on each new task, wasting tool calls on grep and file reads. Graft replaces that with a static graph of linked markdown notes, one per system, API, or concept, explaining the codebase the way a senior engineer would. The structural graph is built deterministically with tree-sitter and never calls a model, while the plain-English summaries can be written by any provider under your own key, including a local model or a LiteLLM proxy. Because the graph is just files on disk, there is no database, server, or hot index to keep running, and the CLI wires it into your chosen agent, refreshing it in the background after each turn. Teams working on large, unfamiliar, or multi-repo codebases who want faster, cheaper agent runs without signing up for a hosted service should find this a practical fit.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.