Fuse documentation
Fuse is a .NET-native codebase context optimizer for AI-assisted development. Start here to install it, connect it to your agent, and learn how it cuts tokens while keeping the public API.
Fuse collects the source files of a .NET codebase, reduces them for token
efficiency, and emits one structured payload an AI agent or a developer can read
in a single call instead of opening thousands of files. It ships as a .NET global
tool (fuse) and as a Model Context Protocol server (fuse serve) with eight
tools.
These docs are organized so you can read top to bottom to learn, or jump to a reference page for exact detail.
Get going
What is Fuse
The outcome, with a worked example and the benchmark figure.
Quickstart
Install, run one command, and read the output.
Connect to your AI
Set up the MCP server for Claude Code, Cursor, or Copilot.
Why Fuse
The problem, the landscape, and how Fuse compares.
The map
- Start: what Fuse is, why it exists, install, the first run, and the MCP setup.
- Scenarios: task-shaped how-tos. Each is a job with one command and real output.
- Concepts: how Fuse works, reduction levels, scoping, the precision tier, and a glossary.
- Reference: every command, flag, MCP tool, reducer, template, tokenizer, and output field.
- Internals and Extending: the pipeline, capability model, scoping and caching internals, and the extension guides.
- Project: benchmarks, performance, roadmap, changelog, and contributing.
Next
New to Fuse? Start with What is Fuse.