Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is actionable with concrete tools and an exact search template, and it sequences the workflow clearly. It loses points for a missing validation/verification checkpoint in a batch operation and for being a monolithic inline document longer than the simple-skill threshold with no progressive disclosure.
Suggestions
Add an explicit validation/verification step for the batch operation — e.g., confirm the number of fetched issue details matches the search results, and handle per-issue fetch failures before building summaries.
Tighten redundant scope statements across 'When To Use' and 'Pass Criteria', or consolidate them, to reduce token cost.
Consider splitting the per-issue output format or reporting rules into a short reference file to apply progressive disclosure, since the body exceeds the simple-skill length threshold.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient procedural guidance with no concept-explanation filler (no 'what is a GitHub issue' padding) and assumes Claude's competence, but sections like 'When To Use' and 'Pass Criteria' restate the same scope, and it could be tightened. It is not score 1 because there is no verbosity explaining known concepts, and not score 3 because of the redundancy and length (~98 lines). | 2 / 3 |
Actionability | Provides a concrete search query template ('is:issue in:title "[Issue]" repo:vladmandic/sdnext'), named tools ('github-pull-request_doSearch', 'github-pull-request_issue_fetch'), and explicit output columns — actionable, copy-ready guidance. This is an instruction-only skill where executable guidance, not code, is the metric, so it earns 3 rather than 2. | 3 / 3 |
Workflow Clarity | A clear four-step sequence (Search, Fetch, Build summary, Produce report) is present, but this is a batch operation over multiple issues with no validation or verification checkpoint (e.g., confirm fetched count vs. matched count, handle fetch errors). Per the rubric, missing validation in batch operations caps workflow clarity at 2; it is not score 1 because the steps are clearly sequenced. | 2 / 3 |
Progressive Disclosure | Sections are well-organized but the skill is monolithic (~98 lines, well over the 50-line simple-skill threshold) with all detail inline and no external reference files (none of references/scripts/assets exist). It is not score 3 because the simple-skills allowance does not apply at this length and no one-level-deep references split the content; not score 1 because organization is reasonable rather than a wall of text. | 2 / 3 |
Total | 9 / 12 Passed |