Enterprise DNA
M MCP Servers Developer low

graphpilot-oss/graphpilot

by Various

Structural memory for coding agents — index your repo once, query symbols, callers, and call-edges over MCP.

G

MCP

graphpilot-oss/graphpilot

Added 11 June 2026

#claude-code #code-analysis #coding-agents #developer-tools #mcp #model-context-protocol #typescript

Overview

An open-source tool that indexes a repository once and exposes structural code information (symbols, callers, call edges) via the Model Context Protocol (MCP). Implemented in TypeScript, it acts as a semantic cache for coding agents to query code relationships without re-parsing.

Best for

Best for
Developers prototyping or building MCP-based coding agents that need structural code context.

Use cases

  • Equip coding agents with a lightweight repo-wide symbol index
  • Query call edges to trace function usage across a codebase
  • Retrieve caller and callee lists for refactoring or dependency analysis

Notes

An open-source tool that indexes a repository once and exposes structural code information (symbols, callers, call edges) via the Model Context Protocol (MCP). Implemented in TypeScript, it acts as a semantic cache for coding agents to query code relationships without re-parsing.

10 stars on GitHub. Last updated 2026-06-10. Licensed Apache-2.0.

Use cases

  • Equip coding agents with a lightweight repo-wide symbol index
  • Query call edges to trace function usage across a codebase
  • Retrieve caller and callee lists for refactoring or dependency analysis

Pros

  • Single-index design reduces repeated parsing overhead for agents
  • Uses standard MCP, making it integrable with MCP-compatible agents
  • Minimal dependency footprint (pure TypeScript)

Cons

  • Very early stage with only 10 GitHub stars and limited community validation
  • No evidence of support for languages beyond the tool’s own implementation language (TypeScript)
  • Requires manual indexing step before queries are possible

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

Pros

  • Single-index design reduces repeated parsing overhead for agents
  • Uses standard MCP, making it integrable with MCP-compatible agents
  • Minimal dependency footprint (pure TypeScript)

Cons

  • Very early stage with only 10 GitHub stars and limited community validation
  • No evidence of support for languages beyond the tool's own implementation language (TypeScript)
  • Requires manual indexing step before queries are possible