Content
88%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 operational, well-sequenced skill body: exact commands, exhaustive exit-code handling, explicit fallback and resume paths, and a strong security section for an untrusted-data crawl. The main costs are repetition of the signature host-binding rule across sections and a single setup reference that is signaled as a bare path rather than a proper link.
Suggestions
State the one-signature-per-host binding rule once (e.g., in 'Hard limits') and cross-reference it from Step 1, exit code 1, and Security instead of restating the full rationale in each place.
Convert the SHOPIFY-SETUP.md pointer into a clearly signaled link (e.g., 'See [SHOPIFY-SETUP.md](extensions/shopify/docs/SHOPIFY-SETUP.md)') so the one-level-deep reference is unambiguous.
Consider moving the exhaustive flag and defaults documentation (concurrency/delay clamps, sitemap walk bounds, precedence rules) into a short reference file, keeping only the flags needed for common crawl decisions inline.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Nearly every line is operational Shopify-specific detail Claude cannot infer (web-bot-auth signature fields, exit-code semantics, clamped defaults), but the one-host signature-binding rule is restated at least four times ('Hard limits', the Step 1 fallback note, exit code 1, and Security) and the Security section overlaps itself on signature handling. This fits anchor 4 (efficient with minor over-explanation that could be trimmed) better than anchor 5, where every token would earn its place. | 4 / 5 |
Actionability | The guidance is fully executable: exact invocations ('"${CLAUDE_PLUGIN_ROOT}/scripts/claude-seo" run shopify_env.py precheck <url>', 'shopify_crawl.py <url> --out ./crawl'), enumerated exit codes with prescribed responses, named fallback reasons, and precise flags and defaults. This matches anchor 5 — copy-paste-ready commands covering the common cases including the signed/unsigned and fallback branches. | 5 / 5 |
Workflow Clarity | The four steps are clearly sequenced (precheck mode decision → crawl → audit on artifacts → score/report) with explicit validation checkpoints: the precheck branches on mode, every crawl exit code has a prescribed recovery action (resume with --resume, fall back to seo-audit, fix the regex pattern), and rate-limit handling distinguishes finished-with-warning from aborted. This is anchor 5 — feedback loops and error-recovery paths are present throughout the batch crawl workflow. | 5 / 5 |
Progressive Disclosure | Good structure with well-labeled sections, a routing table, and a clearly purposed one-level-deep reference ('See extensions/shopify/docs/SHOPIFY-SETUP.md') for setup and env-format details, keeping them out of the main body. It falls short of anchor 5 because the reference is a bare path rather than a clearly signaled link, and the dense flag/defaults documentation could arguably live in a reference file; the bundle itself contains no additional files to verify organization against. | 4 / 5 |
Total | 18 / 20 Passed |