
Codeix
Gives Claude seven read-only tools to query a pre-built semantic index of your codebase without grepping or re-parsing. You commit a .codeindex directory alongside your code, then search for symbols by name or signature, list all functions in a file, find callers or callees of a method, and search comments or docstrings separately from code. Handles Python, TypeScript, Rust, Go, and eight other languages using tree-sitter grammars. Auto-discovers git repos in monorepos and mounts dependency indexes so you can query across package boundaries. Useful when you want Claude to navigate a large codebase structurally instead of burning tokens on trial-and-error file reads.
- GitHub stars
- 9
- tools
- 7
- transport
- STDIO
- last updated
- Feb 21, 2026
Tools
7 toolsPublic tool metadata for what this MCP can expose to an agent.
exploreExplore project structure: metadata, subprojects, files grouped by directory
Explore project structure: metadata, subprojects, files grouped by directory
No parameter schema in public metadata yet.
searchUnified full-text search across symbols, files, and texts (FTS5, BM25-ranked) with scope/kind/path/project filters
Unified full-text search across symbols, files, and texts (FTS5, BM25-ranked) with scope/kind/path/project filters
No parameter schema in public metadata yet.
get_file_symbolsList all symbols in a file
List all symbols in a file
No parameter schema in public metadata yet.
get_childrenGet children of a class/module
Get children of a class/module
No parameter schema in public metadata yet.
get_callersFind all places that call or reference a symbol
Find all places that call or reference a symbol
No parameter schema in public metadata yet.
get_calleesFind all symbols that a function/method calls
Find all symbols that a function/method calls
No parameter schema in public metadata yet.
flush_indexFlush pending index changes to disk
Flush pending index changes to disk
No parameter schema in public metadata yet.
Sponsors
Readme
codeix
codeix.dev · Fast semantic code search for AI agents — find symbols, references, and callers across any codebase.
codeix # start MCP server, watch for changes
codeix build # parse source files, write .codeindex
codeix -r ~/project build # build from a specific directory
Why
AI coding agents spend most of their token budget finding code before they can work on it. They grep, read files, grep again, backtrack. On a large codebase the agent might burn thousands of tokens just locating the right function — or worse, miss it entirely and hallucinate.
Codeix gives the agent a pre-built map of your codebase. One structured query returns the symbol name, file, line range, signature, and parent — no scanning, no guessing.
What existing tools get wrong
Details
- Repository
- github.com/montanetech/codeix
- Website
- codeix.dev
- Category
- Search & Web Crawling
- Package
codeix- Transport
- STDIO
- Registry
- MCP registry, active