Content
86%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 tight, well-structured CLI-wrapper skill: executable safe-start commands, a complete subcommand table, and good layering via the shared parent skill and CLI help/schema discovery. The only gaps are a missing example invocation for the non-'me' subcommands and absent error-recovery guidance for failed auth checks.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and efficient throughout: no explanation of the People API or CLI concepts Claude already knows, a dense command table, and one-line safety bullets — every token earns its place. The only non-instructional line is the generator comment, which is negligible. | 5 / 5 |
Actionability | Copy-paste-ready safe-start commands with concrete flags ('--account user@example.com', '--readonly', '--dry-run') and a full subcommand table, plus explicit contract-discovery commands. Not 5 because only 'me' gets a full example invocation; search/get/relations usage must be obtained via --help/schema rather than shown. | 4 / 5 |
Workflow Clarity | The safe-start sequence (auth check → schema → readonly smoke test) is clearly ordered with checkpoints, and write safety is addressed ('--dry-run before supported writes', 'Confirm the exact account, object, and mutation before any write or delete'). Not 5 because there is no error-recovery feedback loop (e.g., what to do when the auth check fails). | 4 / 5 |
Progressive Disclosure | Well-organized short body with clearly signaled one-level-deep references: '../gog/SKILL.md' for shared rules and runtime 'gog people <command> --help' / 'gog schema people <command> --json' for the machine-readable contract, keeping bulk detail out of SKILL.md. No bundle reference files exist or are needed at this size. | 5 / 5 |
Total | 18 / 20 Passed |