Codegraph
- codegraph-ai/codegraph-vscode
Codegraph is a MCP server that provides 31 semantic graph tools for callers, impact, complexity, and unused code across 15 languages.
About
Codegraph is a semantic code graph exposed as an MCP server with thirty-one tools tuned for how developers actually change code: who calls this, what breaks if I edit it, where is complexity pooling, and what is never referenced. It supports fifteen languages, so repos with polyglot services or scripts still get a unified graph instead of ad-hoc AST grepping. The canonical journey shelf is Ship review because pre-merge impact questions are the killer use case, but the same queries help during backend refactors in Build and when diagnosing failures in Operate. Install through the published npm identifier with stdio transport and point it at your workspace graph. It is analyst tooling for agents, not a formatter or test runner—expect to already have a repo indexed or integrated per upstream docs.
- 31 graph tools for callers, impact, complexity, and unused code
- Semantic code graph across 15 programming languages
- Stdio MCP via @memoryx/codegraph-mcp npm package 0.9.2
- Surfaces change blast radius for agent-driven review
- Pairs with Claude Code or Cursor for navigation without manual ripgrep tours
Codegraph by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add codegraph -- npx -y @memoryx/codegraph-mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| Package | @memoryx/codegraph-mcp |
|---|---|
| Transport | STDIO |
| Auth | None |
| Repository | codegraph-ai/codegraph-vscode ↗ |
What it does
Query a semantic code graph for callers, blast radius, complexity, and dead code before you merge risky changes.
Who is it for?
Best when you're doing agent-assisted code review or refactors on multi-language repos and need graph queries beyond text search.
Skip if: Greenfield prototypes with almost no code history or teams that only need linting without structural analysis.
What you get
After you connect Codegraph, your agent can query semantic relationships and review risk before merges or refactors.
- 31 graph query tools callable from the agent
- Cross-language semantic navigation for review and refactors
By the numbers
- 31 MCP tools
- 15 supported languages
- Server package version 0.9.2
Recommended MCP Servers
0piFree, 2hr expiry agent storage to cache contexts, bridge workflows, or dump data.
100Hires AI ATS & Recruitment SoftwareOfficial 100Hires MCP: AI ATS & Recruitment Software for candidates, jobs, applications, interviews.
123elec McpInterface MCP officielle du site marchand 123elec, catalogue, stock, panier, commandes.
1stayHotel booking MCP server.2
2D Assets MCPMCP server for generating advanced mock 2D PNG assets for game prototypes.43D MeshWeaverMesh spatial optimizer reducing geometric overhead in 3D assets.How it compares
Semantic graph MCP for review, not a cloud deploy or monitoring dashboard.
FAQ
Who is Codegraph for?
Developers and small teams using AI agents who need caller, impact, and complexity intelligence across a real codebase.
When should I use Codegraph?
Use it during Ship review before merging, and during Build or Operate when tracing dependencies or dead code.
How do I add Codegraph to my agent?
Add the @memoryx/codegraph-mcp stdio server from npm to your MCP config following the codegraph-ai repository setup.