Content
82%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 well-structured, token-efficient tool skill whose examples and parameter documentation are immediately usable. The main gaps are the unstated invocation mechanism and the lack of an explicit retry loop for weak results, plus mild redundancy between the two when-not-to-use sections.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean throughout: quick-reference payloads, a parameter table with defaults, a five-step pipeline, and debugging guidance, with no re-teaching of known concepts or padded prose. Nothing stands out as trimmable over-explanation. | 5 / 5 |
Actionability | Copy-paste-ready JSON payloads cover the three main cases (plain prompt, schema, seed URLs) and the parameter table gives types, defaults, and caps, but the invocation mechanism (tool name, CLI, or MCP call) is never stated, leaving a minor executable gap. | 4 / 5 |
Workflow Clarity | "How It Works" gives a clear five-step sequence, and the steps array with "Use steps to debug weak results — if extraction is poor, check which pages were fetched" acts as a checkpoint, but there is no explicit validate-adjust-retry loop for weak results. The batch page fetching is autonomous and non-destructive, so the batch-validation cap does not apply. | 4 / 5 |
Progressive Disclosure | Sections are well organized and the only references are clearly signaled one-level cross-skill links (wigolo-extract, wigolo-research); no bundle files exist to split content into. Minor gaps: boundary guidance is duplicated across "Anti-Patterns" and "When NOT to use wigolo-agent", and the body runs slightly past pure-overview length. | 4 / 5 |
Total | 17 / 20 Passed |