Content
98%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A thorough, well-sequenced mutation workflow with strong validation checkpoints and real, correctly-referenced bundle files. Its main weakness is conciseness — several sections over-explain principles Claude already knows and could be trimmed or moved to a reference.
Suggestions
Trim the Safety & constraints and reply/resolve-ordering sections to only non-obvious rules; remove restatements like 'Never resolve a thread without a reply' that already appear verbatim earlier in step 6.
Move the GraphQL cost-awareness and token-budget guidance into a separate assets/ reference file and link to it, reducing the SKILL.md body length while keeping the detail one click away.
Collapse the long 'When to use' trigger list (which duplicates the description's USE FOR terms) into a brief pointer, since the frontmatter already carries that activation guidance.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is accurate and well-structured but runs long with some padded guidance that Claude could infer (e.g. 'do not assume review feedback is correct', 'never expose secrets', repeated restatements of reply/resolve ordering), so it is mostly efficient but could be tightened. | 3 / 5 |
Actionability | Provides concrete executable commands (real script invocations with flags, GraphQL file paths, `gh api graphql -f query=@assets/...`) with copy-paste-ready examples, but a few steps remain procedural guidance rather than runnable commands. | 4 / 5 |
Workflow Clarity | Clear phased sequence (fetch → analyze → fix → validate → push → reply → resolve → verify) with explicit validation checkpoints, re-fetch-on-error feedback loops, a working checklist document, and a destructive/batch-aware dry-run-first guard for mutations. | 5 / 5 |
Progressive Disclosure | Good structure: SKILL.md overview points one level deep to real bundle files (scripts/*.sh, assets/*.graphql, assets/review-resolution-checklist.md) with a tooling map table; minor gaps such as several large inline sections (cost awareness, token budget) that could live in a reference file. | 4 / 5 |
Total | 16 / 20 Passed |