Blog GEO Citation Audit
Audit whether blog content is ready to be cited by ChatGPT, Perplexity, and Google AI Overviews.
The skill mainly reads content, analyzes structure, and generates a report; no malware, credential theft, or destructive default is evident. However, permission boundaries, user confirmation, sensitive-data handling, rollback, and data-flow disclosure for input content, optional GSC queries, external scripts, and network data are incomplete, so it is far from full marks.
The procedure, scoring tables, and report template are internally consistent and explicitly prohibit fabricated statistics. Repository CI and GEO adapter tests show deterministic outputs, error envelopes, and invalid-input coverage, but they do not directly reproduce the complete Claude Code execution path for this SKILL.md; network failures, missing content, and platform variance remain thinly handled, so the static score is conservative.
Invocation terms, target tasks, file-path input, and declared platform surfaces are fairly clear. Non-fit boundaries, a minimum input contract, Chinese-language handling, and mainland-China network reachability are not specified, and several platform behaviors depend on current documentation, logs, or tooling without an in-skill availability check.
Frontmatter provides a name, trigger description, user-invokable flag, argument hint, and MIT license. The body includes a staged process, scoring rules, limitation disclosures, and a report template, while the README supplies broader installation, version, and CI context. The skill itself lacks versioning, changelog, maintenance ownership, troubleshooting, and dependency notes, and its cross-references depend on surrounding repository context.
For auditing an existing blog post's AI-citation readiness, the procedure, 15-point rubric, 100-point mapping, platform recommendations, and action items can produce a directly usable editorial report. However, the result still depends on reading real content and verifying external sources; no representative end-to-end output is directly evidenced for this skill, and improved citation performance cannot be established statically, so the score remains within the static ceiling.
The skill requires source blocks containing URL, publisher, methodology, sample size, engine/version, query class, retrieval date, and expiry date, and distinguishes heuristic from causal claims. The repository also includes deterministic and error-path tests for GEO adapters. Direct coverage of this SKILL.md's Claude Code report path and independent third-party corroboration are absent, so only limited credit is justified.
- Do not interpret the 0-100 AI Citation Readiness score as a real citation probability or a platform-ranking guarantee.
- Before running optional GSC queries or other network tools, confirm credentials, access scope, and what content may be transmitted; the skill lacks complete data-flow and sensitive-data guidance.
- The core workflow depends on changing platform behavior, crawler policies, and current documentation, without stating mainland-China reachability or offline alternatives.
- Add skill-specific versioning, ownership, changelog, dependency checks, missing-input handling, and end-to-end regression coverage.
What does this skill do, and when should you use it?
Blog GEO is the AI-citation-focused skill in the claude-blog suite, and it does not perform Google ranking optimization. It evaluates passage-level citability, Q&A structure, entity clarity, structured data, and AI crawler accessibility. The skill produces a 0–100 AI Citation Readiness score, platform-specific recommendations, and citation capsules that authors can embed in their articles. It fits writers and content teams optimizing visibility in AI answer and search systems.
Reads a supplied blog file and extracts its word count, heading hierarchy, paragraphs, FAQ sections, schema markup, robots.txt or meta-robots signals, TL;DR boxes, comparison tables, ordered lists, and definition-style formatting. It scores passage-level citability, Q&A formatting, entity clarity, extraction-friendly structure, and AI crawler accessibility, then analyzes likely performance on ChatGPT, Perplexity, and Google AI Overviews. The report includes one 40–60-word citation capsule per H2 section, technical recommendations, priority actions, and a 0–100 overall score.
- A blog author wants a pre-publication check for self-contained passages that AI systems can quote independently.
- An SEO or content team needs to assess whether FAQs, schema, comparison tables, and answer-first sections support AI citation.
- A marketing team wants separate improvement guidance for ChatGPT, Perplexity, and Google AI Overviews.
- An editor needs a scored AI-citation audit with suggested citation-ready passages instead of a full rewrite.
What are this skill's strengths and limitations?
- Its scope is explicit: AI citation readiness rather than combined SEO and ranking optimization.
- Checks concrete factors including passage-level citability, Q&A structure, entity clarity, structured data, and crawler access.
- Provides separate recommendations for ChatGPT, Perplexity, and Google AI Overviews.
- Generates section-specific citation capsules that authors can add to the article.
- It does not optimize Google rankings or perform combined rewriting; the source directs users to `blog-rewrite` for that work.
- The report depends on the supplied article and the technical signals available for inspection; the source does not establish automatic validation of every external site configuration.
- The source provides no dedicated test coverage or measured real-world citation results for this standalone skill.
- Optional GSC analysis requires credentials and Python scripts, so it is not a default capability.
How do you install this skill?
This skill is the skills/blog-geo/SKILL.md component of the claude-blog repository, not a separately documented package. Install the public collection in Claude Code with /plugin marketplace add AgriciDaniel/claude-blog, then /plugin install claude-blog@agricidaniel-claude-blog. Restart Claude Code afterward.
How do you use this skill?
Provide a blog file and run /blog geo path/to/post.md. You can also ask for an “AI citation audit” or “GEO audit” and provide the file path. The skill returns an AI Citation Readiness report. Use /blog analyze <file> separately when you need the collection’s broader content-quality score.
How does this skill compare with similar options?
Unlike blog-rewrite, Blog GEO is limited to AI citation readiness auditing; blog-rewrite is the stated choice for combined Google and AI optimization. Compared with the suite-level /blog analyze, this skill focuses on AI citation rather than the broader 100-point blog-quality rubric.