Content
65%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
Highly actionable body with a real executable command and precise data schemas, but it is somewhat verbose and inlines a long analysis/recommendation playbook that would benefit from being split into reference files, and it lacks validation feedback loops for its batch operation.
Suggestions
Add a post-run verification checkpoint (e.g., confirm the CSV row count matches the requested limit and surface rate-limit/error recovery steps) to lift workflow clarity.
Move the Phase 3 analysis briefing and 7-case recommendation framework into a references/ANALYSIS_PLAYBOOK.md, keeping SKILL.md a lean overview with a clearly signaled one-level-deep link.
Trim restated rationale in the recommendation framework (e.g., "This is the highest-signal segment — call it out explicitly") to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient with concrete commands, scoring weights, and CSV schemas, but the 7-case recommendation playbook and analysis briefing include restated rationale and hand-holding ("call it out explicitly", elaborate cost-estimate breakdowns) that could be tightened. | 2 / 3 |
Actionability | Provides a copy-paste-ready invocation of the real ${CLAUDE_SKILL_DIR}/scripts/gh_repo_signals.py with concrete flags, explicit per-interaction scoring weights, and full CSV column definitions. | 3 / 3 |
Workflow Clarity | Steps 1–8 are clearly sequenced with an environment check, but the long batch profile-fetch operation (~5,000 profiles/hr) lacks an output-verification checkpoint and a rate-limit/error feedback loop, capping workflow clarity at 2. | 2 / 3 |
Progressive Disclosure | Content is well-sectioned and the referenced scripts/gh_repo_signals.py is a real bundle file, but the analysis and recommendation playbook is inlined monolithically with no one-level-deep reference documents (no references/ or assets/ directory). | 2 / 3 |
Total | 9 / 12 Passed |