AC create-release
Cut a new CocoIndex release end to end: propose the next version by semver when the user doesn't give one, draft the release notes from the commits since the last release, create the GitHub release (which creates the tag), and watch the Release workflow publish the wheels to PyPI and deploy the docs, then, for a minor or major release, create its release branch (release/1.2) from the tag. Use whenever the user wants to release, cut, ship or publish a new CocoIndex version: "release 1.0.27", "cut a release", "make a new release", "publish a new version to PyPI", "draft the release", "what should the next version be", "create the release branch for 1.2". A minor or major release is cut from main, and a patch from the release branch of its own minor (1.1.1 from release/1.1); the skill tells them apart.
Cut a new CocoIndex release end to end: propose the next version by semver when the user doesn't give one, draft the release notes from the commits since the…
As a process C 63/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
edit-residuethe text marks something as outdated (lines 234): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 63/100
- 0Result and completion. Does not say what the result is
- 30Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 90 mutating operations with no state check
- 60Tools and files. Uses tools (web, git, python, node) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 4785 tokens
- 100Steps. 30 steps
- 100Failures and branches. 2 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (32 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 810: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 7 example trigger phrases
- +4Structure: 15 headings
- +3Step-by-step instructions: 30 items
- +4Has examples (10 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 89.