skilld
All skills
erkamyaman avatar

/agent-evals

@57c6daa
by erKamerkamyaman/dhh-p-bloom-agent-skills1 stars

Use when measuring how well an AI agent handles a recurring kind of task. Builds a small eval set, runs it, and raises the difficulty once the agent passes.

  • 2 files
  • 2 KB
  • Updated 10 hours ago
  • GitHub

Ask your Agent

Use this Skill: https://skilld.dev/gh/erkamyaman/dhh-p-bloom-agent-skills/agent-evals

This session only. Nothing lands on disk.

referenceseval-template.md

≈135 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Eval template

## Task <id>
Prompt: <exactly what the agent is given>
Starting state: <repo commit or fixture>
Pass check: <command that exits 0 on success>
Difficulty: easy | medium | hard
Notes: <known pitfalls>

Results table

Task Run 1 Run 2 Notes
01 pass pass
02 fail pass flaky check?

Raising the bar

  • Combine two tasks into one.
  • Remove hints from the prompt.
  • Use a larger or messier codebase.
  • Add a performance or size requirement to the pass check.

Source: SKILL.md on GitHub

Third-party checks

No third-party reports yet.

Provenance

Signed by skilld at 57c6daa. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 4 hours ago.

Activeupdated 10 hours ago

README badge

README badge for erkamyaman/dhh-p-bloom-agent-skills/agent-evals