Content
73%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 release workflow with strong validation and good reference structure. Main weaknesses are repeated bare-invocation prose that could be tightened and several referenced scripts missing from the bundle.
Suggestions
Consolidate the repeated bare-invocation intent (currently restated at lines 38-39, 58-60, and 61) into a single canonical statement to reduce redundancy.
Bundle or verify the referenced scripts (release-preflight.sh, check-release-version-sync.py, release-verify-publish.sh, build-opencode.js) so the actionability guidance is fully executable as-is.
Move the full Confirmation Gate plan block and environment-variable tables into a reference file to keep SKILL.md closer to an overview and lift progressive_disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly purposeful dense content, but the bare-invocation intent and 'release complete' framing are restated multiple times (e.g., lines 38-39, 58-60, 61), which is more than minor trimming. | 3 / 5 |
Actionability | Concrete, executable bash/git/gh commands throughout, but several referenced scripts (release-preflight.sh, check-release-version-sync.py, release-verify-publish.sh, build-opencode.js) are not present in the bundle, leaving minor execution gaps. | 4 / 5 |
Workflow Clarity | Explicit 3-stage numbered sequence (Pre-Gate 1-8, Confirmation Gate, Post-Gate 9-19) with abundant validation checkpoints (fail-closed preflight, version-sync, reachability check, verify-publish) and a failure-recovery table with feedback loops. | 5 / 5 |
Progressive Disclosure | Well-signaled one-level-deep references to 6 real references/*.md files (all verified present), but SKILL.md itself is a long ~360-line document that inlines considerable detail beyond a pure overview. | 4 / 5 |
Total | 16 / 20 Passed |