Content
93%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 exemplarily lean, fully executable skill body: concrete CLI invocations for every subcommand, config and environment documented, and useful API caveats (type filter, price levels, color control) with zero padding. The one real gap is that the search/resolve → place_id → details pipeline is never made explicit, leaving the skill's core workflow to inference.
Suggestions
Add one line connecting the steps, e.g. 'search/resolve return place IDs; use one with details --reviews to fetch full info and reviews.'
Briefly note what the human-readable output looks like (fields shown for each subcommand) so Claude knows what to expect and parse.
If goplaces has an --help or subcommand list worth consulting for flags beyond these examples, mention it so Claude can self-serve.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~28-line body is lean: one orienting line ('Human output by default, `--json` for scripts'), then Install/Config/Common commands/Notes with every bullet carrying operational information (flags, env vars, price-level semantics, type-filter caveat). No concept explanations Claude already knows, no padding. Not 4: there is no over-explanation to trim — even the 'Notes' section documents non-obvious API behavior rather than general knowledge. | 5 / 5 |
Actionability | Every example is a copy-paste-ready invocation with real flags: `goplaces search "coffee" --open-now --min-rating 4 --limit 5`, `goplaces resolve "Soho, London" --limit 5`, `goplaces details <place_id> --reviews`, covering the common cases (search with filters, geo-bias, pagination, resolve, details, JSON output). Not 4: the examples cover the full command surface of the tool; the only placeholder, `<place_id>`, is standard and unambiguous. | 5 / 5 |
Workflow Clarity | Each command is unambiguous and this is a simple command-invocation skill with no destructive or batch operations, so no validation checkpoints are needed. Not 5: the primary multi-step workflow — search/resolve returns a place_id, which `details` then consumes — is only implied; the body never states where `<place_id>` comes from or how to chain the commands. Not 3: nothing fragile is involved, and each listed action stands alone and executable. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines with no external references needed, and the body is organized into clear, well-labeled sections (Install, Config, Common commands, Notes) — matching the rubric's simple-skill criterion. No bundle files (references/, scripts/, assets/) exist, so nothing is hidden or nested. Not 4: there is no content that should be split out; the structure is appropriate for the skill's size. | 5 / 5 |
Total | 19 / 20 Passed |