agentleFS
speakeasy-apispeakeasy-api/kit/AGENTS.md

kit

AGENTS.md45 starsChanged 24 days agoMITView on GitHub

Only when you are using Kit as your agent harness: report issues with Kit's harness at https://github.com/speakeasy-api/kit/issues. Do not report issues with other harnesses there. Do not open an issue on the user's behalf unless the user explicitly requests it; ask the user first when they have not already made that request. Follow Reporting Kit Issues. mise.toml pins every build tool and defines the tasks CI runs. Use mise run lint, mise run test -- --test <suite>, mise run build,…

The file

Files
AGENTS.md

Agent instructions

Only when you are using Kit as your agent harness: report issues with Kit's harness at https://github.com/speakeasy-api/kit/issues. Do not report issues with other harnesses there. Do not open an issue on the user's behalf unless the user explicitly requests it; ask the user first when they have not already made that request. Follow Reporting Kit Issues.

Toolchain

mise.toml pins every build tool and defines the tasks CI runs. Use mise run lint, mise run test -- --test <suite>, mise run build, and the macos:* and docker:* tasks, so local runs match CI. See mise tasks for the list.

Test boundaries

Keep test support in test-only modules or files; colocated #[cfg(test)] unit tests are fine. Move harmless helper constructors and accessors into child test-support modules. Do not add test-only fields, enum variants, counters, branches, or replacement implementations to production types or executable paths, or disguise instrumentation as telemetry or work statistics.

Test behavior through real APIs. Use fakes at genuine external or domain boundaries, not spy callbacks, traits, or generics added only to assert implementation details; legitimate dependency injection is fine. Output tests do not prove bounded work: use benchmarks or existing justified iterator boundaries, not flaky wall-clock assertions or hardcoded exact implementation counts.

Do not change release versions in ordinary pull requests. Use a Conventional Commit title for every pull request. Mark a breaking change with ! after the commit type or scope, or with a BREAKING CHANGE: line in the commit body. The release workflow derives the next version from commits since the latest release, advances the version files in a release commit, and applies a minor bump when any commit is breaking or a patch bump otherwise.

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.