CodeCompass cut 66%-82% of token usage on real coding work. And used to map your codebase and reduce the amount of code your agent reads and opens.
CodeCompass cut 66%-82% of token usage on real coding work. And used to map your codebase and reduce the amount of code your agent reads and opens.
A local code knowledge graph that gives AI agents a map of your codebase — so they navigate by structure instead of grepping blind, and know what's connected before they edit. One graph query replaces dozens of file opens: fewer tokens, less compute, and every answer comes with a verifiable file:line. And it compounds — descriptions, edges, and corrections agents write back survive re-indexes and persist across sessions, so each task starts from everything every previous session learned instead of from zero.
Compare MCP servers by category, or browse hosted MCP servers if you prefer an endpoint over a local process.
{ "mcpServers": { "codecompass": { "command": "codecompass-mcp" } } }Configuration as documented by the project. Restart the client after saving.
Read, search and edit files inside allowed directories with the Model Context Protocol reference Filesystem server.
Retrieve version-aware library documentation and code examples with Context7 MCP, using resolve-library-id and query-docs.
An MCP (Model Context Protocol) server for generating O'RLY? (O'Reilly parody) book covers that display directly in Claude Desktop application.
Framework-aware code intelligence MCP server — 60 framework integrations, 81 languages, up to 99% token reduction
A Model Context Protocol server for accessing GitHub repositories containing Obsidian vaults
GitLab MCP Server lets an AI agent (Claude Desktop, Claude Code, Cursor, Zed, VS Code, or any Model Context Protocol client) talk directly to GitLab. It is a typed, paginated, schema-validated…