codebase-memory-mcp-pro — Claude Code skill
codebase-memory-mcp-pro-knowledge-graph is a Claude Code skill that maps a codebase into a syntax-tree graph. Agents find callers, trace call chains, and assess change impact.
no license 100 installs ★ 9 updated 2026-10-09
npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a claude-codeWhat codebase-memory-mcp-pro-knowledge-graph does
codebase-memory-mcp-pro indexes a repository into a knowledge graph with function, class, and cross-file call nodes. This fork fixes incremental re-indexing—editing a file preserves inbound calls to its symbols instead of orphaning them.
The server exposes MCP tools: explore for one-call blast radius with callers and neighbors, query_graph for arbitrary Cypher queries, detect_changes with transitive caller analysis at configurable depth, and trace_path to build call chains between symbols. All results include line numbers. Built from source; the first build compiles 158 tree-sitter grammars and takes a few minutes, but incremental rebuilds are fast.
No precompiled binaries—you build locally to get all integrated fixes in one place.
What it helps with
- find all functions that call a given symbol
- trace a call chain from request entry to database write
- estimate the blast radius of a code change
- identify unreferenced public functions
- explore graph structure with custom Cypher aggregations
Example prompts for the agent
- Show me all callers of
processPaymentand flag hotspots - How many different functions call
handleRequestand where - If I edit
src/payment/processor.ts, which symbols does that impact two hops up the call stack
How to install codebase-memory-mcp-pro-knowledge-graph
1Skills CLI (Vercel)
npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a claude-code2Manually: put the files in place
~/.claude/skills/codebase-memory-mcp-pro-knowledge-graph/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/codebase-memory-mcp-pro-knowledge-graph/
1Skills CLI (Vercel)
npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a codex2Manually: put the files in place
.agents/skills/codebase-memory-mcp-pro-knowledge-graph/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a cursor2Manually: put the files in place
.cursor/skills/codebase-memory-mcp-pro-knowledge-graph/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install reason-machines/mcp-skills skills/codebase-memory-mcp-pro-knowledge-graph2Manually: put the files in place
.github/skills/codebase-memory-mcp-pro-knowledge-graph/SKILL.md1Skills CLI (Vercel)
npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a gemini-cli2Platform CLI
gemini skills install https://github.com/reason-machines/mcp-skills --path skills/codebase-memory-mcp-pro-knowledge-graphManually: put the files in place
.devin/skills/codebase-memory-mcp-pro-knowledge-graph/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about codebase-memory-mcp-pro-knowledge-graph
How do I install codebase-memory-mcp-pro in Claude Code?
Run npx skills add reason-machines/mcp-skills --skill codebase-memory-mcp-pro-knowledge-graph -a claude-code, then open ~/.claude/skills/codebase-memory-mcp-pro-knowledge-graph/SKILL.md.
What does codebase-memory-mcp-pro do?
It parses your repository into a call graph where agents can look up which functions call a given symbol, find paths between symbols, and compute the transitive impact of edits.
Can I use this skill in other AI editors besides Claude Code?
Yes. Install commands are provided for Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode—each with its own platform flag (-a codex, -a cursor, etc.).