codanna
by bartolli·★ 673·Score 53
Codanna is a local code intelligence MCP server providing semantic search, relationship tracking, and document search for AI coding assistants.
Overview
Codanna acts as an X-ray vision for AI coding assistants, enabling them to understand codebases with surgical precision. It provides instant call graphs, semantic search functionality that matches intent rather than just keywords, and full dependency analysis across files. The server is designed for rapid R&D and pair programming situations where traditional LSP tools might be too slow.
Try asking AI
After installing, here are 5 things you can ask your AI assistant:
When to choose this
Choose Codanna when you need fast, semantic code intelligence for AI agents in large codebases and LSP is too slow.
When NOT to choose this
Don't choose Codanna if you need write access to code, real-time collaboration features, or require Windows support which is still experimental.
Tools this server exposes
6 tools extracted from the READMEsemantic_search_with_contextPerform semantic search on code with natural language queries
search_documentsSearch through indexed documentation and text files
call_graphGenerate call graphs for functions to trace relationships
find_implementationsFind all implementations of a function or method
dependency_analysisAnalyze dependencies across the codebase
function_definitionsFind and analyze function definitions
Comparable tools
Installation
Installation
macOS, Linux, WSL
curl -fsSL --proto '=https' --tlsv1.2 https://install.codanna.sh | shHomebrew
brew install codannaWindows (PowerShell)
irm https://raw.githubusercontent.com/bartolli/codanna/main/scripts/install.ps1 | iexSetup for Claude Desktop
Add to Claude Desktop configuration:
{
"mcpServers": {
"codanna": {
"command": "codanna",
"args": ["mcp"]
}
}
}Initialize and Index
codanna init
codanna index srcFAQ
- What programming languages does Codanna support?
- Codanna supports Rust, Python, JavaScript, TypeScript, Java, Kotlin, Go, PHP, C, C++, C#, Clojure, Lua, Swift, and GDScript.
- How does Codanna differ from traditional LSP tools?
- Codanna provides sub-10ms lookups and can parse 75,000+ symbols per second, making it significantly faster than traditional LSP tools for certain operations. It's designed specifically for AI assistants with context-first coding approach.
Compare codanna with
Last updated · Auto-generated from public README + GitHub signals.