Sitemap Inspector & Builder
Analyze or generate valid XML sitemaps and uncover URL, structure, and indexing risks.
The skill lists URL validation, structural checks, location-page warnings, and a 50-page hard stop, reducing obvious misuse risk. It does not define authorization for HTTP requests, external-scope limits, data flows, user confirmation, credential handling, or rollback, so points are deducted.
The two modes, validation items, and abnormal-input responses are clearly described. However, the selected path provides no dedicated script, dependency, or key-path test evidence, and the relative template path may create environment coupling. Static calibration caps this below 10, with further deductions for limited diagnosability and reproducibility.
Trigger phrases, input forms, and analysis/generation scenarios are reasonably clear. Non-fit boundaries, Chinese-language output support, complex-site constraints, and mainland-China reachability are not documented; analysis also depends on access to the target site, so points are deducted.
Front matter supplies name, description, argument hint, MIT license, author, and version. The document is organized by mode, checks, formats, and error handling and includes XML examples. It lacks skill-specific installation/dependency notes, FAQs, changelog, and clear maintenance ownership or update path, so points are deducted.
The checklist, quality signals, generation process, and expected files provide a usable structure for the core task. Static review cannot verify XML generation, HTTP-status checks, lastmod assessment, or report quality, and outputs still require human review, so the score is below the ceiling.
The file gives concrete protocol limits, validation rules, error cases, and reproducible XML examples. It contains no third-party citations, test results, or key-path coverage evidence; repository-level CI material does not establish that this selected skill path was verified, so points are deducted.
- The skill checks HTTP status for arbitrary target URLs but does not document request authorization, rate limits, privacy boundaries, or sensitive-URL handling in this file.
- Generation depends on industry templates under ../seo-plan/assets/; behavior when templates are missing, paths change, or business type is unavailable is unspecified.
- Claims such as all URLs returning HTTP 200 and accurate lastmod values require live network access and human confirmation; static documentation cannot prove execution correctness.
- Chinese-language output and mainland-China reachability are not declared; network restrictions may produce incomplete results when external access is required.
What does this skill do, and when should you use it?
This is the standalone sitemap skill within the claude-seo repository, focused on analyzing existing XML sitemaps or generating new ones. It checks XML validity, URL status codes, file-size limits, canonicalization, noindex and redirect issues, HTTPS usage, robots.txt references, and sitemap-to-crawl coverage. Generation supports industry templates, interactive structure planning, quality gates, and sitemap indexes when URL counts exceed 50,000. It fits technical SEO teams and site owners working on sitemap quality, but the required template path and live-site access should be verified in the execution environment.
In analysis mode, it reads an existing sitemap and validates XML format, URL count, HTTP 200 responses, lastmod quality, deprecated tags, robots.txt references, canonical URLs, noindex URLs, redirects, HTTPS usage, and differences between crawled pages and sitemap URLs. If no sitemap is found, it checks /sitemap.xml, /sitemap_index.xml, and robots.txt references. For unreachable URLs, invalid XML, or rate limiting, it reports status codes, parsing line numbers, or partial results with retry notes. In generation mode, it loads an industry template from ../seo-plan/assets/, plans site structure interactively, applies location-page quality gates, generates sitemap.xml or a sitemap index, and writes STRUCTURE.md; analysis results are written to VALIDATION-REPORT.md.
- An SEO consultant validates a client’s sitemap before launch and identifies redirected, noindex, or non-canonical URLs.
- A technical SEO team audits a large site and decides whether to split sitemaps by content type or use a sitemap index.
- A site owner plans a new site structure and generates an XML sitemap from an industry template.
- A development team investigates malformed XML, inaccessible URLs, inaccurate lastmod values, or missing robots.txt references.
What are this skill's strengths and limitations?
- Covers XML format, HTTP status, URL limits, canonicalization, and indexing-quality signals in one workflow.
- Defines splitting and sitemap-index handling for files exceeding 50,000 URLs.
- Flags noindex, redirected, non-canonical, HTTP, and low-value programmatic page risks.
- Includes a warning at 30 location pages and a hard stop at 50 pages during generation.
- The source does not provide separate test-coverage evidence for this sub-skill.
- Generation depends on industry templates under ../seo-plan/assets/, whose contents are not provided.
- Analysis of URL responses and robots.txt requires access to the target site; no offline alternative is documented.
- Its scope is sitemap analysis and generation, not the full SEO-audit capability of the repository.
How do you install this skill?
There is no separately documented installation for this skill. It is installed as part of the AgriciDaniel/claude-seo collection. For the Claude Code plugin, run /plugin marketplace add AgriciDaniel/claude-seo, then /plugin install claude-seo@agricidaniel-claude-seo. Alternatively, run git clone --depth 1 https://github.com/AgriciDaniel/claude-seo.git followed by bash claude-seo/install.sh.
How do you use this skill?
In Claude Code, run /seo sitemap <url> to analyze a sitemap or /seo sitemap generate to create one. You can also use a request containing “sitemap,” “generate sitemap,” “sitemap issues,” or “XML sitemap.” Analysis produces VALIDATION-REPORT.md; generation produces sitemap.xml, split files and an index when needed, plus STRUCTURE.md.