Content
78%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 is a lean, well-structured instruction skill with concrete evidence-recording requirements, guardrails against fabricating visibility claims, and a clear sequenced workflow ending in re-measurement. Its main room for improvement is per-step validation checkpoints and small concrete examples (e.g., sample queries or a report skeleton), plus a slightly redundant flowchart.
Suggestions
Add per-step validation checkpoints in the prose workflow (e.g., what to confirm before moving from evidence collection to fix proposal), rather than relying on the flowchart's implicit sequence.
Include one or two example audit queries and a minimal report skeleton to make the common case fully copy-paste ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes competence, spending tokens only on non-obvious constraints like 'A crawler allowance or an index file does not establish that an AI service will ingest or cite the site'. The mermaid flowchart partially duplicates the sequence already traceable in the prose, a minor instance that could be trimmed, keeping it just below the every-token-earns-its-place anchor. | 4 / 5 |
Actionability | Guidance is concrete and executable for an instruction-only skill: 'Record the exact query, engine, date, source URL, and observed citation', 'check robots directives, accessible documentation, and structured data', and an explicit three-part report structure. Minor gaps (no example queries or report skeleton) keep it below fully-executable-with-common-cases coverage. | 4 / 5 |
Workflow Clarity | The flowchart lays out a clear five-step sequence (Define → Collect → Separate → Propose → Repeat) and 'Re-run the same dated queries after changes before describing an improvement' is an explicit validation loop; this is not a destructive/batch skill so no cap applies. It falls short of 5 because intermediate checkpoints per step are implicit rather than built into the sequence. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines, single-purpose, needs no external references, and has no bundle files; the body is well organized into three focused phase paragraphs plus a summary flowchart. Per the scoring notes for simple skills, this warrants full marks with well-organized content alone. | 5 / 5 |
Total | 17 / 20 Passed |