Happy slice of mcp pizzaMCP.pizza

FileScopeMCP

MCP.Pizza Chef: admica

It watches a project folder, scores every file by how much else depends on it, tracks which files pull in which, and pulls out functions and classes by name, so your assistant can answer who calls this, what matters most here, or what changed since Tuesday without reading every line. It works with Claude Code, Cursor, and other MCP apps, and adds a browser dashboard for exploring your repositories. No paid key is needed; written file summaries are optional and rely on a model you run locally.

Coding
Data

Use This MCP server To

Show me the most important files in this project Find who calls a function before I rename it Spot circular dependencies I keep tripping over Get a plain summary of what a file does List everything that changed since last Tuesday See which files always get edited together

README

FileScopeMCP

Your AI already knows how to code. Now it knows your codebase.

Build Status Node.js License Trust Score

FileScopeMCP watches your code, ranks every file by importance, maps all dependencies, and keeps AI-generated summaries fresh in the background. When your LLM asks "what does this file do?" — it gets a real answer without reading the source.

Works with Claude Code, Hermes Agent, Codex, OpenClaw, Cursor AI, or as a standalone daemon. Supports TypeScript, JavaScript, Python, C, C++, Rust, Go, Ruby, Lua, Zig, PHP, C#, and Java.

Key Features

Importance ranking — every file scored 0-10 based on how many things depend on it, what it exports, and where it lives. Your LLM sees the critical files first.

Dependency mapping — bidirectional import tracking across all supported languages. AST-level extraction (tree-sitter) for TS/JS, Python, C, C++, and Rust; regex-based for Go, Ruby, Lua, Zig, PHP, C#, and Java. Finds circular dependencies too.

Symbol intelligence — extracts functions, classes, interfaces, types, enums, consts, modules, and structs via tree-sitter for TypeScript, JavaScript, Python, Go, and Ruby. find_symbol resolves names to file + line range. find_callers and find_callees map the call graph for TS/JS so your AI can answer "who calls this function?" before refactoring.

Always fresh — file watcher + semantic change detection means metadata updates automatically. AST-level diffing for TS/JS, LLM-powered analysis for everything else. Only re-processes what actually changed.

LLM broker — a background process coordinates all AI work through llama.cpp's llama-server (or any OpenAI-compatible HTTP API). Priority queue ensures interactive queries beat background processing. Runs on a single GPU.

Nexus dashboard — a web UI at localhost:1234 that lets you visually explore your codebase across all your repos. Interactive dependency graphs, file detail panels, live broker activity, and per-repo health monitoring.

Prerequisites

  • Node.js >= 22 and npm (download)
  • Build tools for native modules (better-sqlite3, tree-sitter):
    • Linux: sudo apt install build-essential python3
    • macOS: xcode-select --install
    • Windows: Visual Studio Build Tools with C++ workload

Quick Start

FileScopeMCP FAQ

Is this actively maintained?
Yes. Code was still landing in May 2026, and the project is being extended rather than wound down.
Which apps does it work in?
Claude Code, Cursor, and other MCP-capable agent tools. The included build script registers it with Claude Code for you, and there are config snippets for the rest.
Do I need to pay for anything?
No. Nothing here calls a paid service. The optional plain-English file summaries need a model running on your own machine, which is also free but takes effort to set up.
Can I use this to understand a project I just inherited?
Yes — that is its best trick. It ranks files by importance and maps what depends on what, so you can start with the parts that actually matter.
Which programming languages does it handle?
It reads TypeScript, JavaScript, Python, C, C++, Rust, Go, Ruby, Lua, Zig, PHP, C#, and Java, with deeper analysis for the first handful.
How hard is the setup?
This one is for developers. You download the project and run a build script, and you need a recent Node install plus your platform's build tools first.
Does it keep itself up to date as I work?
Yes. It watches for file changes and only reprocesses what actually moved, so the map stays current without you rerunning anything.