菜单
Use when cataloging a site, building a page inventory, crawling a sitemap, or listing all pages. Structured URL inventory with SEO signals and page-type labels from sitemaps or domain discovery.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | Site-Content-Catalog |
| type | capability |
| description | Use when cataloging a site, building a page inventory, crawling a sitemap, or listing all pages. Structured URL inventory with SEO signals and page-type labels from sitemaps or domain discovery. |
| version | 1.2.1 |
Produce a structured catalog — URL inventory plus optional SEO signals and page_type labels. Foundational input for playbooks and audits.
Not this skill: cornerstone scoring, cannibalization, or change recommendations.
Reference: REFERENCE.md · Requirements: REQUIREMENTS.md · Examples: EXAMPLES.md
Sitemap URL or domain — from the user only. Never infer from workspace, examples, or prior runs.
Scope (optional) — path prefix such as /blog/ or /features/.
If missing, ask once:
Which domain or sitemap URL should I catalog? (Optional: limit to a path prefix.)
| Mode | When | Data |
|---|---|---|
| sitemap_only | Sitemap + WebFetch; no MCP | URLs, lastmod, page_type; signals null |
| sitemap_enriched | Sitemap + DataForSEO MCP (default full catalog) | + keywords, ETV, backlinks, word/heading counts when available |
| labs_fallback | No sitemap; domain + MCP | Ranked URLs only; limitations[] required |
Completion criterion for mode: handoff.mode matches actual discovery path; limitations[] states any batch cap or partial enrichment.
REFERENCE.md).page_type via classify-page-type.mjs rules (do not invent types).REFERENCE.md).Offer to save handoff JSON for Cornerstone Content Audit or other downstream skills.
Sitemap URL: WebFetch XML; parse <loc>, <lastmod>, <priority>; recurse sitemap indexes.
WordPress / Yoast / core (mandatory when index children match):
scripts/wordpress-sitemap.mjs (parseChildSitemapPostType, shouldIncludeChildSitemap).post-sitemap (posts) and page-sitemap (pages), plus custom CPT sitemaps (e.g. tips-sitemap.xml).*_cat, *_tag) per script defaults.sitemap_post_type from its source child sitemap.discovery_summary.content_sitemaps[] when discovery_source is sitemap_index (one row per included child: sitemap_url, post_type, url_count after dedupe — each url_count must match pages[] rows with that sitemap_post_type; see G19).Domain only: try /sitemap.xml and /sitemap_index.xml; if missing → labs_fallback mode with dataforseo_labs_google_keywords_for_site.
Apply scope filter after discovery.
Completion criterion: every in-scope URL listed once (dedupe www/non-www trailing slashes per REFERENCE.md). WordPress indexes must not omit post (or core wp-sitemap-posts-post-*) child sitemaps when present in the index.
| Signal | Source | Mode |
|---|---|---|
| Ranked keywords / ETV | dataforseo_labs_google_ranked_keywords | enriched |
| Backlinks | backlinks_summary | enriched |
| Title / H1 / depth | on_page_content_parsing or WebFetch | enriched |
| Last modified | sitemap or headers | all |
Large sites (REFERENCE.md): batch enrichment; document cap in limitations[].
Completion criterion: each page has enrichment_status (discovered_only | enriched | unfetchable); enriched fields null only when discovered_only.
Run scripts/classify-page-type.mjs rules (REFERENCE.md). Pass sitemap_post_type when known from WordPress child sitemaps. First match wins.
Completion criterion: every page has page_type from the enum; type_summary counts match pages.length. Flat-permalink posts from post sitemaps must be blog_post, not landing_page.
## Site Content Catalog — {domain} — {date}
**Mode:** {mode} · **Discovered:** X · **Enriched:** Y · **Scope:** {all | prefix}
**Discovery:** {sitemap index children included — e.g. post, page, tips — or single sitemap}
| URL | Title | Type | Sitemap PT | Top Keyword | ETV | Backlinks | Last Modified |
| ... |
Emit handoff_version: "1.0" per REFERENCE.md. When a playbook calls this skill, return handoff directly (playbook formats the report).
All modes
page_type and enrichment_statustype_summary sums to pages.lengthlimitations[] lists batch caps, labs fallback, or missing v1.0 fieldsnode scripts/verify-handoff.mjs (G1–G19)node scripts/verify-scorecard.mjs (T1–T15, W1–W9)sitemap_enriched additionally
enriched_count reflects pages with MCP signalsdiscovered_only rowsPlaybook consumer: Cornerstone Content Audit Phase 1 — do not assume inbound_internal_links (not in v1.0; see limitations).
After classification or handoff schema changes:
node scripts/verify-scorecard.mjs
node scripts/verify-handoff.mjs
Use when analyzing fan-out coverage, sub-query coverage, topic coverage gaps, or keyword-only SERP fan-out without a domain. Simulates AI query fan-out and optional domain coverage checks for a keyword.
Use when creating or reviewing an agent skill, adapting an external skill for public release, reviewing SKILL.md quality, or choosing handoff JSON and verify scripts. Right-sized packaging with predictable process; regression gates only when needed.
Use when auditing keyword cannibalization, same-domain page overlap, or hub-vs-hub conflicts before consolidation. Use for accidental hub vs intended hub, blog post outranking a landing page, or internal competition diluting rankings on one domain.
Use when auditing robots.txt, AI crawler permissions, GEO crawl policy, cornerstone path crawlability, or blocking/allowing GPTBot, PerplexityBot, OAI-SearchBot, and Google-Extended. Audits selective crawler rules per bot (training vs indexing crawl), validates Sitemap directives, and drafts policy-aligned robots.txt.
Use when recovering hub-and-spoke content architecture on a messy site — accidental hubs, hidden gems, legacy bloat, or onboarding a client without a designed pillar map. Use for organic architecture discovery, de facto pillar audit, GSC and GA4 triangulation, or bottom-up cornerstone recovery. Triggers: "hub and spoke audit", "discover my real pillars", "accidental cornerstone", "hidden gem pages", "performance matrix content audit", "triangulate GSC and GA4 with keyword data".
Use when running a cornerstone content audit, pillar content review, or content architecture audit. End-to-end: discover pages, identify cornerstones, health audit, cannibalization check, and enhancement plan.