SEO Content Brief Generator
Builds actionable SEO content briefs from competitor results and the target website’s real structure.
The skill instructs the agent to fetch the target page, homepage, sitemap, and SERPs, but does not define request permissions, data flows, sensitive-data handling, user confirmation, or rollback. It does include a stop-and-report rule for unreachable URLs, and repository-level security material describes safeguards, but external-request scope and execution boundaries remain incomplete.
The process, two operating modes, output structure, and abnormal-input table are broadly coherent, with supporting references for domain filtering, templates, and keyword rules. However, this path has no dedicated tests or key-path reproduction evidence, and its character-limit wording is not fully consistent, so the static score remains moderate-low.
Trigger phrases, argument hint, new/improve modes, page-type handling, and expected output are reasonably clear. It does not specify Chinese-language output or mainland-China network fit, and core research depends on access to target sites, SERPs, and optional external services; non-fit boundaries are therefore incomplete.
The skill has YAML metadata, an MIT license, layered process documentation, reference files, templates, error handling, and a fixed output format. It lacks skill-specific maintenance ownership, changelog, update path, installation/dependency notes, and FAQs; governance evidence is mostly repository-level, so points are deducted.
It can produce a directly usable brief covering competitors, gaps, outline, metadata, E-E-A-T, and internal links, while enforcing website relevance and information-gain requirements. However, competitor and keyword outputs depend on live retrieval, factual accuracy requires human review, and no verifiable representative result for this path is supplied; the static ceiling limits the score.
The instructions call for research-backed analysis, citations, and real competitor data, while repository-level README and CI provide limited indirect support. The selected path and its references contain no concrete source URLs, dedicated tests, or independent reproduction record, and fact-versus-inference separation is underspecified, so only limited credit is justified.
- The skill performs external retrieval against user-supplied sites and search results; confirm scope before auditing untrusted targets and do not provide credentials or sensitive URLs.
- Rules such as keyword density and title/meta character ranges are presented as hard SEO requirements without verifiable sources in the selected materials; validate them before publication and avoid mechanical application.
- Chinese-language briefs, Chinese SERPs, and mainland-China-reachable data sources are not addressed, so additional research channels may be required for Chinese users.
What does this skill do, and when should you use it?
This is the content-brief skill inside the Claude SEO repository, focused on planning new pages and improving existing ones. It combines the target site, sitemap, and search results to score competitors, identify content gaps, and shape the page structure. Its briefs cover keyword placement, section word counts, metadata, information gain, E-E-A-T signals, and internal linking. It fits SEO consultants, content leads, and site operators who need research-backed writing plans.
Reads a target URL or keyword, the target website’s homepage and sitemap, and the top ranking pages for the query. It filters out non-competitors, scores remaining pages for depth, formatting, SEO, and UX, and identifies topic, depth, and quality gaps. It classifies search intent and the SERP format being rewarded, then produces a page-type-specific outline with section word counts, format notes, featured-snippet targets, and keyword guidance. It also generates title and meta-description recommendations, a specific information-gain requirement, E-E-A-T signals, and 3–5 internal-link opportunities. Optional DataForSEO and Ahrefs MCP integrations can add live research data, but they are not required for the core workflow.
- An SEO consultant is planning a new service page and needs competitor research for the target query.
- A content lead is improving an existing article and needs separate keep, strengthen, and add-new recommendations.
- A site operator is building a hub or category page and must cover real products, services, or child pages in the site structure.
- A business writer needs a handoff-ready brief with word counts, keyword placement, metadata, and trust-signal requirements.
What are this skill's strengths and limitations?
- Supports both new-page briefs and improve-existing-page briefs.
- Uses website relevance and site structure to constrain recommendations and find internal-link targets.
- Produces competitor scoring, section word counts, keyword guidance, metadata, and a concrete information-gain requirement.
- The core workflow does not require paid data services; DataForSEO and Ahrefs are optional integrations.
- The core research process depends on access to the target site, sitemap, and search results; it cannot be fully completed without network access.
- Missing sitemaps can make internal-link recommendations incomplete.
- The source provides no test count, accuracy measure, or outcome evidence for this individual skill.
- The source does not document installation or invocation for clients other than Claude Code.
How do you install this skill?
Install the full Claude SEO repository because this skill is one of its 33 bundled skills. With Claude Code 1.0.33+, 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. The source does not document a standalone installation procedure for this skill.
How do you use this skill?
In Claude Code, run /seo content-brief <topic>, such as /seo content-brief "commercial roofing services". You can provide an existing page URL and page type to use improve mode. Natural-language triggers include “content brief,” “write a brief,” “content outline,” “blog brief,” “service page brief,” “content plan,” and “outline for.” To request only an outline, say “just an outline” or “content outline.”