Content
71%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, phase-by-phase workflow that is largely actionable and reasonably token-efficient, with concrete commands, regex heuristics, and an output template. Its main weaknesses are the absence of any bundle files (everything is inlined in one long SKILL.md) and missing error-recovery guidance for crawl failures or empty sitemaps, plus minor duplication between the When to Use and Trigger Phrases sections.
Suggestions
Move the full report template (Phase 6) and the cost tables into references/ files (e.g. references/report-template.md, references/costs.md) and link to them from SKILL.md, keeping the body as an overview.
Add a validation/recovery step after Phase 1 for sitemap crawl failures or competitors with few URLs (e.g., 'If sitemap returns <100 URLs, fall back to internal crawl from the homepage; if a competitor sitemap is unreachable, note it and continue'), which would raise workflow clarity.
Deduplicate 'When to Use' and 'Trigger Phrases' (they list near-identical phrases) and consolidate the env-var configuration into the single Tools Required section.
Replace the Phase 4 placeholder comment with an actual fetch_webpage invocation, and complete the DataForSEO example with a runnable request (auth header, curl command) so the enhanced-mode path is copy-paste ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational — regex clustering tables, per-phase commands, and a cost table — with only minor redundancy such as 'When to Use' and 'Trigger Phrases' duplicating the same phrases and env-var lists appearing in both the Tool Enhancement section and Tools Required. Not 5: these duplicated sections could be trimmed; not 3: the padding is minor and most tokens carry operational value. | 4 / 5 |
Actionability | Provides an executable catalog command ('python3 skills/site-content-catalog/scripts/catalog_content.py --url ... --output json'), a concrete regex table, and a full report template, but the DataForSEO call is a bare endpoint fragment rather than a runnable script and Phase 4 shows only a placeholder comment ('# Pick pages from the pattern') instead of a fetch command. Not 5: these gaps keep it from copy-paste ready throughout; not 3: most guidance is concrete and executable. | 4 / 5 |
Workflow Clarity | A clear Phase 0–6 sequence with explicit intake, mode selection, and per-phase outputs, plus check steps in baseline mode ('spot-check 3-5 pages per pattern via web_search', 'Check if pages are indexed'). Not 5: there are no explicit error-recovery or validation feedback loops (e.g., what to do when a sitemap crawl fails or returns few URLs); not 3: the sequence is complete with most checkpoints present, and the read-only analysis means the destructive/batch cap does not apply. | 4 / 5 |
Progressive Disclosure | The skill is a single ~275-line file with no references/, scripts/, or assets/ — the full output report template, the tool-setup script, and the cost tables are inlined where separate reference files would keep SKILL.md an overview. Not 4: substantial content that belongs in separate files is inline and the >50-line simple-skill exception does not apply; not 2: the body is well-sectioned with clear headers, so it is navigable despite being monolithic. | 3 / 5 |
Total | 15 / 20 Passed |