Content
92%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 tightly written, highly actionable skill: executable script commands with expected output, explicit validation gates with error recovery, and an exemplary references table with per-reference when-to-use and skip conditions. The only trimmable content is the mildly rhetorical intro and Mindset framing.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and directive with no explanations of concepts Claude already knows, but the framing intro ('Write down the one thing this session is for, keep it honest as work lands') and the rhetorical 'Mindset' section ('Trade convenience for provability everywhere') could be trimmed without losing guidance, fitting the efficient-with-minor-trimming anchor. | 4 / 5 |
Actionability | Fully executable, copy-paste-ready commands — `./scripts/goal.sh new "..."`, `park 3 "..."`, `status`, `check` — with verbatim console output, concrete YAML status fields, and a BAD/GOOD evidence table covering the common cases. | 5 / 5 |
Workflow Clarity | The sequence (clarity test → record → set goal-status → evidence → park → summary → promote → close) is clear and gated by explicit validation ('Close out only when `check` exits zero'), includes an error-recovery example with a FAILED message and exit code, and ends with a verification checklist. | 5 / 5 |
Progressive Disclosure | The ~130-line overview keeps detail in five real bundle files (all verified present), surfaced through a References table that gives each one a 'When to Use' and a skip condition; references are one level deep with only benign sibling cross-links, making navigation easy. | 5 / 5 |
Total | 19 / 20 Passed |