Skip to content
Get packs

Build and Test Runner

Runs your project's typecheck, lint and test commands and reports whether the build passes.

Source Download skill
Repository
nexu-io/open-design
Stars
105,448
Licence
Apache-2.0
Files
2 files

What it does

This skill makes the agent run your project's own build, typecheck and test commands after it edits code, rather than relying only on a subjective review. It records whether the build and tests passed, how long they took, and the captured output, so a larger automated loop can keep fixing code until checks succeed or a retry limit is reached.

Great for

  • Developers doing code migrations or automated refactors who need proof the build still works
  • Teams using the open-design plugin pipeline with a defined test command for Next, Vite or Remix projects

Skip it if

  • Projects without a configured typecheck or test script, since the skill depends on those commands existing in package.json or being supplied as input
  • Anyone wanting it to judge design quality or code style, because it only checks exit codes of commands, and running arbitrary project commands can be risky on untrusted repositories

Try asking

  • “Run the build and tests after the migration step and report whether they pass”
  • “Use pnpm test as the test command and keep retrying fixes until the build passes or eight iterations are reached”
  • “Show me the build log from the last verification run”
Step-by-step guide: how to use Build and Test RunnerInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the Apache-2.0 licence and an attribution note.

    build-test.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip build-test.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip build-test.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Set an explicit testCommand input when the project's scripts are non-standard
  • Check critique/build-test.log when results look wrong, since output is truncated at 1 MiB by default
  • Review any skipped tests, which the skill requires to include a reason

What's inside

  • open-design.json
  • SKILL.md

Redistributed unmodified from nexu-io/open-design under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.