Recon MCP Server

Lightweight code intelligence engine for AI agents. Dependency mapping, blast radius analysis, and call graph via MCP.

Local serverstdioPython

What is the Recon MCP server?

Recon MCP server exists for a simple reason — assistants are far more useful when they can act on Recon directly instead of describing what you should do. Lightweight code intelligence engine for AI agents. Dependency mapping, blast radius analysis, and call graph via MCP.

What you get

Give your AI agent a brain. Index your codebase in 5 seconds.

What the assistant can call

Once Recon is connected, these are the calls the assistant has available:

  • Method — Path
  • GET — /api/health
  • POST — /api/tools/:name
  • recon_map — Architecture overview: packages, tech stack, entry points, health
  • recon_find — Smart search: exact name, wildcard (*Handler), or natural language
  • recon_explain — Full 360° context: callers, callees, flows, cross-language links, tests
  • recon_impact — The recon_impact tool exposed by this server
  • recon_changes — Git diff to affected symbols, risk assessment, and affected tests
  • recon_rename — The recon_rename tool exposed by this server
  • recon_export — The recon_export tool exposed by this server
  • recon_rules — Code quality: dead code, circular deps, unused exports, large files, orphans

Setting it up

Installation goes through your MCP client rather than a global install: point it at recon-mcp on npm and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.

Choosing this one

Among the developer tooling options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Recon's toolset — Method, GET, POST and 8 more — is a fair guide to whether it matches your workflow. It is maintained by jhm1909; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Before you rely on it

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 11 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Recon.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the recon mcp server does with a few real requests.

Available tools

ToolWhat it does
MethodPath
GET/api/health
POST/api/tools/:name
recon_mapArchitecture overview: packages, tech stack, entry points, health.
recon_findSmart search: exact name, wildcard (*Handler), or natural language.
recon_explainFull 360° context: callers, callees, flows, cross-language links, tests.
recon_impactThe recon_impact tool exposed by this server.
recon_changesGit diff to affected symbols, risk assessment, and affected tests.
recon_renameThe recon_rename tool exposed by this server.
recon_exportThe recon_export tool exposed by this server.
recon_rulesCode quality: dead code, circular deps, unused exports, large files, orphans.

How to install the Recon MCP server

```json
{
  "mcpServers": {
    "recon": {
      "command": "npx",
      "args": ["recon-mcp", "serve"],
      "cwd": "/path/to/backend"
    }
  }
}

Configuration as documented by the project. Restart the client after saving.

Example prompts to try

  • Use Recon to Method.
  • Use Recon to GET.
  • Use Recon to POST.

Frequently asked questions

It connects Recon to MCP-compatible AI assistants such as Claude and Cursor, exposing 11 tools (Method, GET, POST, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Recon directly.