MCP Tools DirectoryAbout this integration
Intelligent code knowledge graph for AI coding agents — 71% cheaper, 72% faster
- Transport
- stdio
- Authentication
- none
- Initial setup
- none
- Runtime
- unattended
- Evidence
- documented
- Version
- 0.1.2
- Package
- @pyalwin/codemesh
- Last compatibility test
- Not independently tested
Connect your agent
@pyalwin/codemeshPublisher documentation was reviewed from the pinned README and package manifest. The graph can read and index a configured local project and can persist agent write-back. Optional embedding setup may acquire a local model. The package was not executed or independently security-audited.
Capabilities: Index code symbols, imports, calls, and git relationships, Search code and assemble relevant context, Trace call chains and reverse dependencies, Persist agent summaries and workflow paths
Connected profiles
Additional details
io.github.pyalwin/codemesh
Source ↗ · Checked 2026-09-170.1.2
Source ↗ · Checked 2026-09-17CC0-1.0; package licenses are separate
Source ↗ · Checked 2026-09-17npm
Source ↗ · Checked 2026-09-17@pyalwin/codemesh
Source ↗ · Checked 2026-09-190.1.2
Source ↗ · Checked 2026-09-19Publisher documentation describes a persistent local MCP knowledge graph, a package command launched by the client, and a local stdio entry point; this review did not run it.
Source ↗ · Checked 2026-09-19The documented MCP configuration supplies only CODEMESH_PROJECT_ROOT, and the documented optional local embedding index is described as zero API cost; no service credential or user authorization is documented for local MCP operation.
Source ↗ · Checked 2026-09-19The publisher explicitly documents a stdio transport that launches npx with @pyalwin/codemesh and sets CODEMESH_PROJECT_ROOT to the local project.
Source ↗ · Checked 2026-09-19Publisher documentation reviewed; package and integration endpoint not executed or independently security-audited.
Source ↗ · Checked 2026-09-19