agentleFS
Sign inStart free
steerlabssteerlabs/opensteer/.agents/skills/run-agent-evals/SKILL.md

run-agent-evals

Skill204 starsChanged 7 days agoMITView on GitHub

Use when running agent tool-use behavioral evaluations (`pnpm eval:agent-tools`), checking a model-facing tool, guide or skill change, repeating the personal memory cases, or comparing a treatment with a compatible baseline.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Stop everything you start by captured PID or serial, even after a failure.

    line 14
  2. Very specific:

    Never drive the host screen: GUI and Electron checks run only in the VM (pnpm test:vm).

    line 15
  3. Distinctive:

    Never read or operate signing keys, Apple accounts or release credentials.

    line 16
  4. Distinctive:

    Never use a person's real profile, credentials, saved passwords, Messages or personal stores.

    line 17
  5. Very specific:

    Never touch the live install: ~/Library/Application Support/Opensteer, ~/Opensteer, Opensteer Beta.

    line 18
  6. Distinctive:

    Use an existing subscription login, add no API keys, and never target host files in a case.

    line 21

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Run agent evals

Real-engine evals spend local subscription capacity and run by hand. Read the safety boundaries first.

Safety rules

  • Devices have one owner: never take another task's simulator, emulator, USB phone or VM slot.
  • Stop everything you start by captured PID or serial, even after a failure; never kill by name.
  • Never drive the host screen: GUI and Electron checks run only in the VM (pnpm test:vm).
  • Never read or operate signing keys, Apple accounts or release credentials.
  • Never use a person's real profile, credentials, saved passwords, Messages or personal stores.
  • Never touch the live install: ~/Library/Application Support/Opensteer, ~/Opensteer, Opensteer Beta.
  • Skipped, blocked or unrun checks are unverified, never passed.
  • Use an existing subscription login, add no API keys, and never target host files in a case.

Procedures

  • Run and compare: harness tests, focused runs, baselines.

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.

Check your setup, free