Content
75%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 operational skill: concrete commands, a clear five-step checklist, and an error-recovery table. The main gaps are the missing worked example for actor input JSON, no explicit post-run validation step, and a script dependency that is not part of the visible bundle.
Suggestions
Include one concrete example JSON_INPUT for the most common case (e.g., a compass/Google-Maps-Reviews-Scraper query with a placeId and maxReviews), since constructing actor inputs is the step most likely to fail.
Add an explicit validation checkpoint between Steps 4 and 5: confirm the output file exists and contains rows (or the chat results are non-empty) before summarizing, with a retry path if empty.
Verify the referenced ${CLAUDE_PLUGIN_ROOT}/reference/scripts/run_actor.js actually ships with the skill bundle, or inline the script's usage contract (flags, exit codes) so the dependency is self-contained.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean: an actor-selection table, exact commands, and a terse error table with no explanation of concepts Claude already knows. Minor trims are possible — the H1 paragraph repeats the description verbatim, "(No need to check it upfront)" is filler, and "Based on character of use case" is vague — so it does not reach the every-token-earns-its-place bar of 5. | 4 / 5 |
Actionability | Steps 2 and 4 provide concrete, mostly copy-paste bash commands with clearly marked placeholders (ACTOR_ID, JSON_INPUT) for all three output modes. It falls short of 5 because no concrete example JSON_INPUT for a common case (e.g., a Google Maps reviews query) is given, leaving the most error-prone step — constructing the actor input — without a worked example. | 4 / 5 |
Workflow Clarity | The five steps are clearly sequenced with a copyable progress checklist, and the Error Handling section provides error-to-fix feedback ("Run FAILED" → check console link, "Timeout" → reduce input or raise --timeout). It is not 5 because there is no explicit post-run validation checkpoint (e.g., verify the output file exists and is non-empty before reporting) and the mcpc schema output is not checked before building the input. | 4 / 5 |
Progressive Disclosure | The body is well-organized into sections that each earn their place, and it appropriately delegates actor input schemas to a runtime fetch rather than inlining them. It is not 5 because it depends on ${CLAUDE_PLUGIN_ROOT}/reference/scripts/run_actor.js, which is not present in this skill's bundle, and the large actor table is inline content that could live in a reference file. | 4 / 5 |
Total | 16 / 20 Passed |