Skip to content
MCP server

Code Crawler

Trace what really runs in Java, Python, TypeScript and JavaScript, with edge confidence

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.1.92
Listing updated
3 days ago

About Code Crawler

Code Crawler is an open-source MCP server for web search and scraping. It runs locally on your machine and is started by your AI client.

Plain grep finds strings, but questions like who transitively calls a method or what breaks if an interface changes are graph traversals. Code-Crawler indexes a project once into a single SQLite file and answers them with recursive SQL, with no external service. The site mentions 29 MCP tools.

Its graph includes the things static call graphs miss: interface and base-class calls fan out to every override, injected dependencies, AspectJ and Spring AOP advice, reflective lookups, and routes or listeners invoked by a framework. Every edge carries a confidence so an assistant can tell fact from guess.

Re-crawls use a per-file manifest of size, modified time and SHA-256, and the site cites 0.9 s versus 22 s on a 137k-line project. Parsing uses tree-sitter WASM grammars.

Install

Run in a terminal

claude mcp add code-crawler -- npx -y @nithin554/code-crawler

Add to claude_desktop_config.json

{
  "mcpServers": {
    "code-crawler": {
      "command": "npx",
      "args": [
        "-y",
        "@nithin554/code-crawler"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "code-crawler": {
      "command": "npx",
      "args": [
        "-y",
        "@nithin554/code-crawler"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "code-crawler": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@nithin554/code-crawler"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "code-crawler": {
      "command": "npx",
      "args": [
        "-y",
        "@nithin554/code-crawler"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "code-crawler": {
      "command": "npx",
      "args": [
        "-y",
        "@nithin554/code-crawler"
      ]
    }
  }
}

Configuration

  • CODECRAWLER_DBPath to the SQLite graph database (default ~/.code-crawler/index.db)
  • CODECRAWLER_EXCLUDE_DIRSComma-separated directories to skip during crawl
  • CODECRAWLER_DEFAULT_LIMITDefault result cap for list queries (default 50)
  • CODECRAWLER_MAX_DEPTHTraversal depth cap (default 10)
  • CODECRAWLER_HELPFUL_ERRORSFuzzy suggestions on unknown symbols (default true)

Tools this server provides (13)

Taken from the project documentation; the running server may differ.

crawl_project
Indexes a project into the graph.
search_symbols
Searches symbols.
get_callers
Finds callers of a symbol.
get_callees
Finds callees of a symbol.
trace_path
Finds a path between two symbols.
trace_execution
Traces execution from a symbol.
find_implementations
Finds implementations of an interface or base class.
find_routes
Lists routes and endpoints.
find_injections
Lists injected dependencies.
find_aspects
Lists aspect advice.
find_hotspots
Ranks heavily connected symbols.
find_dead_code
Finds dead-code candidates.
get_decorators
Queries decorators and annotations.

What it does

Call tracing
Finds transitive callers, callees and shortest paths with call-site lines.
Virtual dispatch
Expands interface calls to every override.
Runtime wiring
Models routes, dependency injection, reflection and AOP advice.
Hotspots and dead code
Ranks by fan-in, fan-out and advice count.
Incremental crawls
Re-parses only changed files.

Good for

  • Find everything that calls a method
  • Judge the blast radius of an interface change
  • List endpoints and what they call
  • Spot dead-code candidates

Pricing

Open source

Prices not published; runs locally.

Prices change. Check nithin554.github.io before you buy.

Code Crawler FAQ

Which languages are supported?

Java, Python, TypeScript and JavaScript.

Where is the index stored?

In a single SQLite file, index.db, by default under ~/.code-crawler.

Does it need a server?

No server, docker or external database.

Can I skip folders?

Yes, with the CODECRAWLER_EXCLUDE_DIRS setting.

Is Code Crawler free?

Yes. Code Crawler is open source, so you can use and self-host it for free.

How do I add Code Crawler to Claude Code?

Run this command in your terminal: claude mcp add code-crawler -- npx -y @nithin554/code-crawler

What are the best alternatives to Code Crawler?

The closest alternatives to Code Crawler are REA, Scrapling MCP Server, Codebase Memory, GitHub and Serena MCP: the IDE for your agent.

Alternatives to Code Crawler

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. MCP server for AI clients to scrape sites with HTTP and stealth browser sessions

    Runs locally
    • 17 tools
    • Open source
    • 104K GitHub stars
    • Web search and scraping
  3. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  4. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  5. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  6. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory