Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →ai-creed/{{name}}
TODO(shakespii):
Passed 22 of 22 test cases with the skill, and 8 of 22 without it.
and 11 more in the full report
14 fixed8 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
I need to document a new automated deployment capability for our team. Could you scaffold a complete skill definition file with all standard markdown sections like intent, inputs, procedure, and anti-patterns so I can fill in the details?
| Checked | Without skill | With skill |
|---|---|---|
| Includes the section heading '## Intent'. | Missed | Passed |
| Uses unnumbered H2 headers for standard sections. | Missed | Passed |
| Places the output contract section before the examples section. | Missed | Passed |
The instructions your model reads when the skill loads.
TODO(shakespii): one or two sentences — what capability this skill provides and why it exists.
TODO(shakespii): what the skill consumes — arguments, files, context. Mark which are optional.
TODO(shakespii): every external dependency — binaries on PATH, env vars, network access, files, path-layout assumptions. If it can fail before step 1, it belongs here.
TODO(shakespii): numbered steps. Calibrate freedom — prose for judgment steps, exact commands for fragile ones.
TODO(shakespii): the deliverable's shape and contract, resolvable from files inside this skill directory — never "obey the handoff format" pointing at unshipped documents.
TODO(shakespii): at least one concrete input→output worked example. Trigger-phrase lists do not count.
TODO(shakespii): common mistakes, rationalizations, and when NOT to use this skill.
test cases passed, vs 8 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.