Content
80%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 concise, well-structured meta-skill body with concrete fallback commands, but the workflow lacks validation checkpoints and the fallback omits a command for surfacing new comments/issues.
Suggestions
Add a verification step before persisting (e.g., confirm the digest is non-empty / within max_words before calling memory_save) to introduce a feedback checkpoint.
Include a concrete gh command or filter for new comments/issues (e.g., 'gh pr view --comments' or a search query) so the fallback fully covers the three buckets.
Make the primary pull → digest → memorize sequence explicit in the body rather than relying solely on frontmatter composition.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean — 'pulls open PRs / failing CI / new comments via the `github` skill (gh CLI), digests into three actionable buckets, and records follow-ups to long-term memory' — with no over-explanation; every token earns its place. | 5 / 5 |
Actionability | The fallback gives concrete executable commands ('gh pr list', 'gh run list --status failure', 'memory_save'), but a command for fetching new comments/issues is missing and 'summarize' is left implicit. | 4 / 5 |
Workflow Clarity | The pull → digest → memorize sequence is present (in frontmatter composition plus the body summary) but there are no validation checkpoints, which caps this batch-aggregating workflow at 3. | 3 / 5 |
Progressive Disclosure | This is a sub-50-line skill with no external references needed; the body is well-organized into a concise overview and a clearly labeled Fallback section, meeting the simple-skill exception. | 5 / 5 |
Total | 17 / 20 Passed |