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.
The body is an information-dense, highly actionable operations guide: a concrete discovery command, a precise mapping of checks to script output fields, explicit manual validation steps, and complete XML templates. Weaknesses are minor — some repetition of the 50k/50MB limits, an underspecified generation mode that points at non-bundled template assets, and extension-sitemap detail that could be split into a reference file.
Suggestions
Deduplicate the 50,000-URL/50MB limits: state them once in the checks table and have the Common Issues table and Process step 6 reference that row instead of restating them.
Make the generation mode executable: either bundle or inline a concrete example of an industry template from the referenced assets directory, and specify what 'interactive structure planning' concretely involves.
Add an explicit verification step after 'Generate valid XML output' (e.g. re-run sitemap_discovery.py or a lint check on the generated file) so the generation workflow has the same feedback loop as analysis.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational — check-to-output-field mapping tables, complete XML templates, and no padding explaining what sitemaps are. Minor trimming is possible: the 50k-URL/50MB limits are repeated in the checks table, the Common Issues table, and Process step 6, which keeps it at anchor 4 rather than 5. | 4 / 5 |
Actionability | A copy-paste-ready command with flags ("scripts/claude-seo run sitemap_discovery.py <url> --expand --json"), an exact output-field mapping per check, and complete valid XML examples make the analysis path fully executable. Anchor 5 is missed because generation-mode guidance is less concrete: "Load industry template from ../seo-plan/assets/" points at assets not in this bundle and "Interactive structure planning with user" is underspecified. | 4 / 5 |
Workflow Clarity | Mode 1 has an ordered discovery process with an explicit feedback loop ("truncated, children.skipped; rerun with a higher --max-children") and by-hand validation checks; Mode 2 has numbered steps with quality-gate checkpoints (⚠️ at 30+, 🛑 HARD STOP at 50+). It stays at 4 because step 5 "Generate valid XML output" has no explicit verify-the-output step, and the operations are read-only/non-destructive so the 3-cap does not apply. | 4 / 5 |
Progressive Disclosure | Sections are well-organized with clear headers and tables, and there are no nested references. At ~175 lines with everything inlined, the detailed image/video/news extension rules are a plausible candidate for a separate reference file, and the referenced script/templates live outside the bundle, which leaves minor organization gaps at anchor 4. | 4 / 5 |
Total | 16 / 20 Passed |