Skip to main content
Glama

graphdex

The code graph that's never stale.

A Python library + CLI + MCP server that maintains an always-correct structural code graph (functions, classes, imports, calls, tests) and serves token-budgeted, confidence-labeled context to AI coding agents.

Status: v0.1 core engine in development — Python parsing, two-pass import-aware resolution, subtoken search, and CLI are implemented; freshness (self-healing reads) and the MCP server land next. Design spec: docs/superpowers/specs/2026-07-17-graphdex-design.md.

Why graphdex

Code-graph tools for AI agents share four recurring weaknesses:

Problem elsewhere

graphdex answer

Stale graphs served as fresh when watchers die

Self-healing reads — every query dirty-checks the working tree and repairs the graph inline; every response carries a freshness field

Cross-file calls matched by bare name (false blast radii)

Import-aware two-pass resolution with per-edge confidence tiers (RESOLVED / IMPORT_INFERRED / NAME_ONLY) and min_confidence filtering

30 MCP tools taxing every agent request

8 consolidated tools that return source snippets inline within a token_budget

Search that only sees exact identifier names

Subtoken FTS5 indexing — getUserById matches "user by id"

Related MCP server: OpenCodeHub MCP Server

Planned surfaces (v1)

  • Library: from graphdex import Graphdex; gd = Graphdex("."); gd.callers_of("login")

  • CLI: graphdex build | status | search | impact | serve

  • MCP: graphdex serve (stdio) — works with Claude Code, Cursor, Codex, and any MCP client

v1 languages: Python, TypeScript, JavaScript (+JSX/TSX), Go, Rust, Java.

Credits

Clean-room implementation inspired by the excellent code-review-graph (MIT) — our analysis of its trade-offs shaped this design. No code is copied from it.

License

MIT

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Provides AI assistants with a structured, token-efficient map of a codebase's symbols, dependencies, and relationships via MCP tools like overview, query, and impact analysis.
      9
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Provides code intelligence for AI coding agents by indexing repositories into a hybrid knowledge graph, enabling agents to query dependencies, impact, and context through 28 MCP tools.
      4
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      Serves structured code context via MCP, enabling AI agents to understand codebases with dependency graphs and significantly reduce token usage.
      10 npm
      1
      MIT