Content
67%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.
The body delivers a genuinely actionable workflow: concrete actor IDs, executable commands for all three output modes, and a specific error-recovery table. Weaknesses are localized: a filler 'When to Use' section and other noise hurt conciseness, the run_actor.js helper's interface is under-documented, and validation is not integrated as in-step checkpoints.
Suggestions
Delete the boilerplate 'When to Use' section and the '(No need to check it upfront)' parenthetical; replace vague 'Based on character of use case' with concrete default guidance (e.g., suggest 20-50 results for a quick scan).
Document run_actor.js's supported flags (including --timeout) and include one real --input example (e.g., a hashtag or keyword object) so the JSON_INPUT placeholder is concrete.
Move the 19-row actor catalog into a references/actors.md file and add an explicit checkpoint in Step 2 (e.g., confirm the schema fetch returned parameters before proceeding to Step 4).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The commands, actor table, and error-handling list are tight, but there is clear slack: the boilerplate "When to Use" section ("Use this skill when tackling tasks related to its primary domain or functionality as described above") says nothing, the "(No need to check it upfront)" parenthetical and "Based on character of use case" are noise, and the "Best For" column largely duplicates the "User Need" column. This fits anchor 3 (mostly efficient but some unnecessary content that could be tightened) better than 4, where only minor trimming would be needed. | 3 / 5 |
Actionability | The mcpc schema-fetch command and all three run_actor.js invocations are concrete and executable with clearly explained placeholders, and the error-handling section gives specific remediations. Minor gaps keep it below anchor 5: no example of a real JSON_INPUT payload, and the --timeout flag is referenced in error handling but never shown in any usage line. | 4 / 5 |
Workflow Clarity | A clear five-step sequence with a copyable progress checklist, explicit step boundaries, and a dedicated error-recovery section covering the main failure modes. Not 5 because validation is not woven into the steps as explicit checkpoints (e.g., verify the schema fetch succeeded before running the actor) and Step 5's reporting is a loose list rather than a verification gate; not 3 because the sequence and error feedback are well beyond 'present but checkpoint-less'. | 4 / 5 |
Progressive Disclosure | The body is a well-organized single file with clean section headers, and the actor catalog plus commands are reasonably placed. It misses anchor 5 because the referenced helper script (${CLAUDE_PLUGIN_ROOT}/reference/scripts/run_actor.js) is not part of this bundle and its interface is undocumented, and the 19-row actor catalog is exactly the kind of lookup material that belongs in a separate one-level-deep reference file. | 4 / 5 |
Total | 15 / 20 Passed |