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 highly actionable, well-organized skill body with concrete commands, examples, and honest error/partial-data guidance. The main weakness is the tangential self-hosting section, which inflates token cost without aiding normal use.
Suggestions
Trim or relocate the self-hosting section (Docker/PostgreSQL/Meilisearch setup and the resource-heavy mirror caveat) to a separate reference file, since it is not needed to run search/detail against the hosted API.
Consider moving the full facet-vocabulary reference (region/country/seniority/category values) into a bundled references file with a one-line pointer, keeping SKILL.md as a lean overview.
Add a short explicit 'validate the result' checkpoint (e.g. check meta.count and stderr error code before acting on results) to nudge workflow_clarity toward the top anchor.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The command, flag, example, and partial-data sections are efficient and API-specific, but the ~25-line self-hosting section (Docker Compose, PostgreSQL, Meilisearch, and the resource-heavy mirror caveat) is tangential to using a search CLI and could be tightened or moved out. | 3 / 5 |
Actionability | Fully executable copy-paste CLI commands with a complete flag reference and six concrete examples covering common cases (senior backend, remote React EU, DevOps Germany, ML/AI, plain-text descriptions, detail lookup). | 5 / 5 |
Workflow Clarity | Clear search-vs-detail sequencing with explicit error-handling checkpoints (429/5xx backoff, unreachable API exits non-zero, 404 reported as an error not an empty set); as a read-only skill it has no destructive/batch cap, but it lacks a formal validate-fix-retry loop. | 4 / 5 |
Progressive Disclosure | No bundle files exist, so all content lives in one well-sectioned SKILL.md; structure is good with no nested references, but at ~183 lines with reference-style facet vocabularies inlined it is a touch above the simple-skill threshold rather than an overview pointing to split materials. | 4 / 5 |
Total | 16 / 20 Passed |