cocoindex-code vs Graft
Graft is much bigger: 5.6k stars against 2.7k. Over the days we have tracked them cocoindex-code moved +49.3% and Graft +1199.3%, so Graft is growing faster right now.
Neither one leads on the six capability axes, so the choice comes down to which of them fits the way you already work.
Stars and commit dates come from our own daily tracking. The six axes are read off each project's documentation by our review pipeline, so they describe what a project says about itself, not what we measured in its code.
Where they stand today
AST-based semantic search that is extremely lightweight, agent-friendly (skill/MCP plugin), and can run fully locally with offline embeddings.
- Stars
- 2.7k
- Tracked growth
- +49.3%
- Maturity
- ●●●●●
- Last commit
- 3d ago
- Language
- Python
- License
- Apache-2.0
- Cost to run
- Local embeddings free with the 'full' variant; slim/latest uses cloud embeddings and your API key.
Persists a deterministic, human-readable context graph as markdown files inside the repo (no embeddings or external index), so agents can follow real files and share an up-to-date code understanding via git.
- Stars
- 5.6k
- Tracked growth
- +1199.3%
- Maturity
- ●●●●●
- Last commit
- 1d ago
- Language
- TypeScript
- License
- MIT
- Cost to run
- Your API key; provider billing applies.
Six axes, head to head
Each axis runs 0 to 5. The label under a score is what that project's own docs claim, not a category average.
| Axis | cocoindex-code | Graft |
|---|---|---|
Context depth How much of your codebase it sees before it answers: the open diff, the diff plus related files, or the whole repository. | ●●●●● Whole-repo analysis | ●●●●● Whole-repo graph |
Noise control How it keeps output volume down — severity thresholds, deduplication, incremental runs over new commits only. | ●●●●● Path/lang filters | ●●●●● Ranking & checks |
Customization How far it bends to your team: custom rules, prompts, style guides, per-path config. | ●●●●● Config file options | ●●●●● Config & flags |
Privacy Whether your code stays on your own infrastructure: fully local, self-hostable, or cloud API only. | ●●●●● Fully local run | ●●●●● Self-hostable |
Model freedom Whether you can point it at any provider, or it is wired to one. | ●●●●● BYO key / local | ●●●●● Bring-your-key |
Setup ease What it takes to get a first useful run out of it. | ●●●●● One‑minute setup | ●●●●● One-command init |
Which one to pick
Pick cocoindex-code if…
Easy setup — install in ~1 minute with zero config and optional local embeddings for offline/private use.
Pick Graft if…
Easy setup — one-command init builds a local, versionable code-graph and wires directly into coding agents (Claude Code et al.) to cut tokens, tool calls, and latency.
What people want from each one
cocoindex-io/cocoindex-code
Hacker News: Show HN: CocoIndex-Code open source embedded AST-based code MCP drew 2 points and 1 comments.
NanoNets/Graft
Questions people ask
Is cocoindex-code better than Graft?
Neither one leads on the six capability axes, so the choice comes down to which of them fits the way you already work. cocoindex-code is worth picking when easy setup — install in ~1 minute with zero config and optional local embeddings for offline/private use.
Which of cocoindex-code and Graft keeps my code private?
cocoindex-code: Fully local run (5/5). Graft: Self-hostable (4/5).
What does each one cost to run?
cocoindex-code: Local embeddings free with the 'full' variant; slim/latest uses cloud embeddings and your API key.. Graft: Your API key; provider billing applies..
Full profiles: cocoindex-io/cocoindex-code and NanoNets/Graft. Everything else in Codebase context.