flyto-indexer is a zero-dependency MCP server that gives AI agents deep understanding of your codebase — impact analysis, cross-project references, and code quality scoring.

The Problem

AI coding assistants are great at writing code, but they don't understand the blast radius of their changes. Modifying a shared utility function might break 15 call sites across 3 projects — and the AI has no way to know.

The Solution

flyto-indexer indexes your codebase and provides 30 MCP tools for code intelligence:

pip install flyto-indexer
flyto-index scan /path/to/project

Impact Analysis

Before changing any function, check what depends on it:

> impact_analysis("flyto-core:src/utils/format.py:function:format_date")

Risk: HIGH
Call sites: 23
Projects affected: 3 (flyto-core, flyto-cloud, flyto-ai)

Dead Code Detection

Find code that nobody calls:

> find_dead_code(project="flyto-core", min_lines=10)

Found 12 unreferenced functions (total 340 lines)

Code Health Score

Get a quick quality overview:

> code_health_score(project="flyto-core")

Score: 82/100 (Grade: B)
- Complexity: 85/100
- Dead code: 78/100
- Documentation: 80/100
- Modularity: 86/100

Supported Languages

Python (AST-based), TypeScript, JavaScript, Vue, Go, Rust, and Java.

Get Started

Read the full flyto-indexer documentation or install it now:

pip install flyto-indexer