agentleFS
Litenova-SolutionsLitenova-Solutions/Fuse/site/llms.txt

Fuse / site

llms.txt37 starsChanged 18 days agoApache-2.0View on GitHub

Fuse is a .NET tool that makes dotnet build and dotnet test faster for AI coding agents. It keeps a repository's C# solution loaded in a background process, the engine. It reports only the compiler errors an edit introduced, including breaks in other projects. It runs only the affected tests. It prints only build errors. All agents and subagents in a repository share one engine: their checks and test plans run at the same time, each answered about its own…

The file

Files
llms.txt
# Fuse

> Fuse is a .NET tool that makes `dotnet build` and `dotnet test` faster for AI coding agents. It keeps a repository's C# solution loaded in a background process, the engine. It reports only the compiler errors an edit introduced, including breaks in other projects. It runs only the affected tests. It prints only build errors. All agents and subagents in a repository share one engine: their checks and test plans run at the same time, each answered about its own edits, and their builds take turns. Agents use it through hooks or an MCP server. It runs locally and has no telemetry. Its main cost is memory. Once idle, it uses 319 MB to 512 MB on the benchmark repositories, up to 1,091 MB while it checks, and more on a machine with more memory.

Fuse needs the .NET 10 SDK. Install it with `dotnet tool install -g Fuse`. Then run `fuse init` in a repository. `fuse init` registers Fuse's hooks with each harness the repository uses (Claude Code, Cursor, Gemini CLI, Codex, GitHub Copilot CLI, OpenCode and Muse Code), and the MCP server with VS Code agent mode. After each edit, the agent gets only the errors that edit introduced, compared with HEAD, the commit you have checked out. When the agent runs `dotnet test`, only the affected tests run and only failures are printed. Fuse is open source under the Apache-2.0 license.

## Docs

- [Overview](https://fuse.codes/docs/): what Fuse is, what it gives an agent, what it costs, the words the documentation uses, and which page answers which question
- [Getting started](https://fuse.codes/docs/getting-started): install Fuse, set it up in your repository, and see what it reports
- [Connect your agent](https://fuse.codes/docs/harnesses): what `fuse init` writes for each harness, and how an MCP host runs `fuse mcp`
- [Subagents and parallel agents](https://fuse.codes/docs/parallel-agents): several agents in one repository, what each is told, `fuse init --instructions` and `fuse stop`
- [How it works](https://fuse.codes/docs/how-it-works): how Fuse checks an edit, selects tests and runs them
- [Limits](https://fuse.codes/docs/limits): what Fuse cannot check, and what to do about it
- [Benchmarks](https://fuse.codes/docs/benchmarks): benchmarks on four repositories, with the result file behind each number
- [FAQ](https://fuse.codes/docs/faq): why Fuse exists, common questions, and how Fuse differs from language servers, Roslyn MCP servers, code search tools and test impact analysis tools
- [Commands](https://fuse.codes/docs/commands): every `fuse` command, its arguments, its output and its exit codes
- [Messages and fixes](https://fuse.codes/docs/messages): every message Fuse prints when it cannot answer, and the fix
- [Troubleshooting](https://fuse.codes/docs/troubleshooting): silent hooks, slow checks, log locations and how to stop the engine
- [Changelog](https://fuse.codes/docs/changelog): what changed in each release, the upgrade steps, and which versions are supported

## Optional

- [Architecture](https://fuse.codes/docs/architecture): how the code is organized, for contributors and coding agents
- [Evals](https://fuse.codes/docs/evals): how the evals work and how to run them
- [AI policy](https://fuse.codes/docs/ai-policy): how you may use AI tools to contribute
- [Contributing](https://raw.githubusercontent.com/Litenova-Solutions/Fuse/main/CONTRIBUTING.md): how to build, test and submit a change
- [Security](https://raw.githubusercontent.com/Litenova-Solutions/Fuse/main/SECURITY.md): supported versions and how to report a vulnerability
- [GitHub repository](https://github.com/Litenova-Solutions/Fuse): source code, issues and releases
- [NuGet package](https://www.nuget.org/packages/Fuse): the Fuse dotnet tool package
- [Eval results](https://github.com/Litenova-Solutions/Fuse/tree/main/evals/results): the JSON result files behind every published number

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.