# kunal12203/GrapeRoot

Compounding Context for AI Coding Assistants — MCP graph engine for Claude Code, Cursor, Copilot, Gemini, OpenCode

Repository: https://github.com/kunal12203/GrapeRoot
Canonical: https://ross.abutalabs.com/products/graperoot
Homepage: https://graperoot.dev
Language: PowerShell
License: Apache-2.0
License Family: permissive
Topics: ai-coding, claude-code, context-engine, copilot, cursor, gemini, mcp, graperoot
Last push: 2026-09-02T07:38:04+00:00

## Health v2 (maintenance only)
Score: 73/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 72, longevity 13
- inputs: {"age_days": 193, "days_push": 0, "days_rel": 186, "gap_med": 0, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1033, forks 123 (observed 2026-09-03T02:15:14.216677+00:00)

## What it is
GrapeRoot is an open-source context engine and MCP server that builds a semantic graph of a codebase (files, symbols, imports, call chains) and pre-loads relevant code into AI coding assistant prompts. It runs 100% locally and integrates with Claude Code, Cursor, Copilot, Codex, Gemini, and OpenCode to reduce token waste and exploration turns.

## Use cases
- reduce token usage when using Claude Code on a large codebase
- give my AI coding assistant better codebase context
- stop my AI assistant from re-exploring files every turn
- index my project so Cursor or Copilot finds the right files
- run a local MCP server that maps my codebase
- track which files my AI assistant read and edited in a session
- speed up multi-turn AI coding sessions on a monorepo

## When to choose
- you use AI coding assistants (Claude Code, Cursor, Copilot, Codex, Gemini) on medium-to-large codebases
- you want to cut token costs and exploration overhead in AI coding sessions
- you need a fully local, privacy-preserving context layer with no cloud dependency
- you want zero-config setup that auto-connects to your AI tool

## When to avoid
- you don't use AI coding assistants or MCP-compatible tools
- you need cross-repo, branch-aware, or org-scale features available only in the paid tiers
- your project uses languages outside its supported parser set
- you prefer your AI assistant to explore the codebase itself rather than receive pre-loaded context

## Facets
- artifact type: cli-tool
- maturity: active
- function: mcp, search-engine, rag, developer-tools, parser
- domain: developer-tools, large-language-models, programming-languages
- platform: windows, cli, python
- tags: context-engine, mcp-server, code-graph, token-optimization, ai-coding-assistants, semantic-graph, local-first, claude-code, cursor, copilot, ai-agents, macos, linux, nodejs

## Member repositories
- kunal12203/GrapeRoot (main) score 73

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:14.216677+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:07:47.219528+00:00, confidence not recorded.
  - readme: https://github.com/kunal12203/GrapeRoot (fetched 2026-09-03T02:15:14.216677+00:00, sha 3bbe7238a1f5)
  - homepage: https://graperoot.dev (fetched 2026-08-29T13:07:44.388732+00:00, sha 4596df5f139a)
  - site_page: https://graperoot.dev/about (fetched 2026-08-29T13:07:44.391501+00:00, sha c6707cef8fdb)
  - site_page: https://graperoot.dev/docs (fetched 2026-08-29T13:07:44.395226+00:00, sha 0608b02e65bc)
  - site_page: https://graperoot.dev/pricing (fetched 2026-08-29T13:07:44.393477+00:00, sha a26cfb236cb4)
- Data as of 2026-08-30T08:39:29.467469+00:00.