Content
100%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.
An exemplary instruction-only skill body: dense and lean, unambiguous about what to do at each step, with real validation and error-recovery checkpoints, and clean delegation of shared detail to sibling skills. The batch operation (multiple Slack threads, PR merge) is fully covered by verification steps, so no cap applies.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every sentence carries an operational instruction ("Open every screenshot, file, video, transcript, and linked recording; the error text there is often the whole diagnosis") with no padding and no explanation of concepts Claude already knows. | 5 / 5 |
Actionability | Highly concrete, executable instruction throughout: "add 👀 before investigating, one thread at a time", "Timebox one item to about 30 minutes", "run /ship through the merge, with every fixed thread's Slack link in the PR body's Reports section", plus a worked classification example ("Objectively broken (errors, dead clicks, lost or wrong data, regressions...): fix it"). | 5 / 5 |
Workflow Clarity | Six clearly sequenced stages (read → decide → clusters → fix/verify/ship → close the loop → recap) with explicit validation checkpoints and feedback loops: "verify as verifying-changes describes (a test that fails before and passes after)", "If the reaction fails, resolve that before going on", "confirm the reply landed under the right parent". | 5 / 5 |
Progressive Disclosure | The body inlines only what is specific to working a given thread and delegates the shared contract to named sibling skills ("That skill owns the Slack contract... Read it first and follow it"), with a closing "Related skills" section; structure is clear with one-level-deep references and no bundle files are needed. | 5 / 5 |
Total | 20 / 20 Passed |