Flyto Indexer

Gives coding agents a local code-intelligence map for impact analysis, dependency graphs, and dead code detection before and after edits.

What you can do

Flyto2 Indexer builds a local index of a repository and exposes MCP tools such as search, impact, task, audit, structure, and verify. It shows callers, dependents, likely tests, cross-project links, and unresolved references before an edit, and checks index freshness, policy, secrets, documentation, and working-tree state before an agent declares work finished. It also records task state and feedback locally so work can resume across clients.

When to use it

Useful when an AI coding agent works on an existing codebase where a change can cross packages, services, or repositories, or when several developers or agents must follow the same repository rules. It is aimed at risky refactors and at verifying that requested work, changed paths, and required checks agree before merge.

Requirements

Runs as a local process on the user's machine, installed from PyPI as flyto-indexer (pip install flyto-indexer). No API key, hosted account, or source upload is declared. Desktop only; a CLI and an optional loopback-only HTTP bridge are available.