Agent skill

Write Content

by inhouseseo in inhouseseo/superseo-skills

A skill your agent uses when writing a complete SEO article.

Apache-2.0Auto-check passedWriting & Content

Install Write Content

skills CLI
$ npx skills add inhouseseo/superseo-skills --skill write-content -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install inhouseseo/superseo-skills write-content --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/inhouseseo/superseo-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/write-content .claude/skills/write-content && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
write-content
GitHub stars
350
Token cost
~3.1k tokens
SKILL.md length
1,650 words
Files
35 (incl. references)
Skills in repo
11
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when writing a complete SEO article.

  • Works in 4 steps: Research → Content Type Decision → Knowledge Extraction → …
  • Writing a complete SEO article
  • SKILL.md covers Input, Business context persistence, Phase 1: Research and Phase 2: Content Type Decision, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Write Content is an agent skill from inhouseseo/superseo-skills. Use when writing a complete SEO article. Includes the full anti-AI-slop ruleset (banned vocabulary, banned phrases, banned structural patterns) and voice rules. The agent researches the SERP itself if needed — no keyword data exports required.

Its SKILL.md is about 3.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 36 other files, including reference files (for example `references/anti-slop-ruleset.md`, `references/content-types-overview.md` and `references/content-types/alternatives-page.md`).

It sits in Writing & Content, covering Blog and article writing. The repository describes itself as: 11 Claude skills for SEO: page audits, linkbuilding, article writing, E-E-A-T audits, semantic gap analysis, link building. Methodology from Koray Tuğberk, Kyle Roof, and Lily…. The licence is Apache-2.0.

When your agent uses it

  • Writing a complete SEO article
  • Tasks that involve Blog and article writing

Example prompts

  • “/write-content”

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. Research
  2. Content Type Decision
  3. Knowledge Extraction
  4. Write the Article

What it can do on your machine

Read from SKILL.md and the folder at commit 9cf22cc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Write Content loads about 3.1k tokens when it runs, and up to ~60k if it reads all its reference files. Until then it costs about 64 tokens; SKILL.md has 1,650 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~64
When it runs · the whole SKILL.md, loaded when a task matches
~3.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~60k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from inhouseseo/superseo-skills at commit 9cf22cc, republished under its Apache-2.0 licence (© inhouseseo). 1,650 words, ~3,136 tokens.

Download SKILL.mdSave it as .claude/skills/write-content/SKILL.md (or your agent's skills folder). This skill also uses 34 other files; get the full folder from GitHub.
name
write-content
description
Use when writing a complete SEO article. Includes the full anti-AI-slop ruleset (banned vocabulary, banned phrases, banned structural patterns) and voice rules. The agent researches the SERP itself if needed — no keyword data exports required.

Write Content

Writes a complete SEO-optimized article. Four phases: research → content type decision → knowledge extraction → write. Includes the full anti-AI-slop ruleset and the voice rules that make the output sound like a practitioner, not a press release.

Input

  • Topic or target keyword (required)
  • (Optional) An existing content brief — skip the research phase if provided
  • (Optional) Expert interview output from the expert-interview skill

If no topic is given, ask for one before proceeding.

Business context persistence

Business context (audience, tone, language, brand voice, examples) shapes every article. Don't re-ask these questions every session.

First use: ask 4-5 questions and save the answers somewhere persistent in your agent's environment. Use the safest default first:

  • Claude Code: ~/.claude/projects/<path>/memory/business-context.md is the recommended default. Do NOT write to ./CLAUDE.md unless the user explicitly asks for it — CLAUDE.md is the user's project instructions file and appending unsolicited content to it can surprise them on every subsequent agent turn.
  • Claude Desktop / Claude.ai: save to a Project's context or a pinned note.
  • Cursor: .cursor/rules/business-context.md.
  • Any other agent: seo-context.md in the working directory.

Always confirm the write location with the user before saving. If any of these are unavailable or the user objects, fall back to re-asking the questions each session.

Every use after that: load that file first. If missing, ask where the user saved it or re-ask the questions.

Questions on first use:

  • What does the business do, and who is it for?
  • What's the brand's tone of voice? (Professional / Casual / Technical / Authoritative / Conversational)
  • What language should content be written in?
  • What topics should NEVER appear? (compliance, competitor mentions, etc.)
  • Who are the 2-3 main competitors?

Phase 1: Research

If a content brief wasn't provided, Google the topic and read the top 5 results. Note: what formats are ranking, what angles exist, what gaps you see. 3-5 bullet points, not a full brief.

Skip this phase entirely if a brief or prior conversation context already contains SERP analysis.

Phase 2: Content Type Decision

Based on what's ranking, pick a content type: how-to, definition/explainer, comparison (X vs Y), listicle/roundup, product review, case study, pillar/ultimate guide, FAQ, landing page, service page, news/trend analysis.

State it plainly: "The top results for [keyword] are all [format]. I'll write a [content type] with [key structural element]. Sound good, or did you have something else in mind?"

Wait for confirmation. If the user already asked for the article in one go, or there's nobody to respond (autonomous run), state your choice with a one-line reason and keep going.

Load references/content-types-overview.md for the decision table covering all 23 content types. Then load the specific template from references/content-types/<type>.md (e.g., references/content-types/how-to.md) for H1/H2 structure, schema, featured snippet format, CTA placement, word count targets. The 19 content types bundled as full templates: how-to, definition, comparison, listicle, pillar-page, faq-page, landing-page, service-page, case-study, statistics-page, news-article, glossary-page, alternatives-page, buying-guide, product-page, category-page, integration-page, location-page, programmatic-page. For the 4 types covered only by the overview table (thought-leadership, product-reviews, pricing-pages, about-pages), those live under eeat-audit/references/content-types/ because the E-E-A-T bar for them is the load-bearing factor.

Phase 3: Knowledge Extraction

Ask 2-3 quick questions to extract unique knowledge the user has. Pick from:

  • "What do most people get wrong about [topic]?"
  • "Can you give me a specific example — a client, a project, a number?"
  • "What surprised you when you actually did this?"
  • "Who should NOT follow this advice, and why?"

Ask one at a time. Keep it quick.

If the user can't or won't answer — autonomous run, or they skip the questions — write from Phase 1 research alone and note in the delivery where first-party input would lift the article.

Adapt style:

  • Newer/smaller site, less SEO-savvy user: conversational, explain why each question matters
  • Established site, experienced user: fast, direct, no hand-holding

Phase 4: Write the Article

Length: Do not target a specific word count. Match the depth of top-ranking content from Phase 1. Length follows intent and competition — never pad to hit a number.

Produce the complete article in clean markdown. Follow ALL of these rules:

Voice and Stance
  • Write like a practitioner talking to a peer. Not a textbook, not a press release.
  • Take clear positions. "We tested this and X works better than Y" beats "both X and Y have merits."
  • Use "you" and "I/we" — write to one person, not an audience.
  • Include specific numbers, names, dates. Never "many companies" — always "[Company] in [year]."
  • Specifics must be real: pulled from the Phase 1 research, the interview answers, or the business context. Never invent a number, name, study, or citation — an invented specific is worse than a generic sentence. No real one available? Ask, or cut the claim.
  • Weave in interview answers as first-person experience. Preserve phrasing where it sounds natural.
  • Use contractions: "doesn't" not "does not."
  • Show thinking changing: "At first I thought this was a branding problem — turns out it was pricing all along." Self-correction is a human signal.
  • Anchor in real context — reference current events, industry shifts, or cultural touchstones where relevant.
Rhythm and Structure
  • Vary sentence length dramatically. Mix 5-word punches with 30-word complexes. Never 3+ consecutive sentences of similar length.
  • Vary paragraph length. One-sentence paragraphs are fine. So are 6-sentence ones.
  • Use fragments for emphasis. Start sentences with "And" or "But" when natural.
  • Include parenthetical asides and brief tangents — humans do this, AI doesn't.
  • Shift registers. After a technical explanation, drop into a casual aside. Uniform register = AI tell.
  • Break the topic-sentence-support pattern. Start some paragraphs with an example, a question, or a statement that only makes sense after reading on.
  • Cover sections asymmetrically. Spend 500 words on the interesting part and 50 on the boring-but-necessary one.
  • Don't summarize at the end of sections unless genuinely complex (3+ subsections).
Show, Don't Just State
  • Don't state facts. Show them through brief scenarios. Instead of "page speed affects rankings" — "You click a search result. Three seconds pass. Still loading. You hit back. Google tracked every millisecond."
  • For claims backed by experience, narrate the moment: what was tried, what happened, what surprised you.
Show full SKILL.md (658 more words)Show less
Anti-Slop Rules

NEVER use these words — highest-signal AI tells: delve, landscape (metaphorical), testament, leverage, utilize, robust, seamless, furthermore, moreover, additionally, pivotal, multifaceted, harness, embark, navigate (metaphorical), showcase, streamline, paramount, culminate, spearhead, commence, endeavor, vibrant, innovative, comprehensive (as adjective).

NEVER use these phrases: "It's worth noting", "In today's [anything]", "Let's dive in", "In conclusion", "plays a crucial/vital/pivotal role", "It goes without saying", "In the realm of".

Avoid these structural patterns:

  • Rule-of-three groupings (use 2 or 4 items instead)
  • Synonym cycling (repeat the right word rather than finding alternatives)
  • Copula avoidance ("serves as" — just say "is")
  • Em-dash chains (max 1-2 per 1000 words)
  • Binary contrasts ("it's not X, it's Y" — just make the argument)
  • Participial tack-ons ("...highlighting the importance of X" — delete or make a separate sentence)
  • Clustering of: however, notably, essentially, that said, arguably — fine individually, but 3+ in one article flags AI
Content Type Structure
  • How-to: 40-60 word quick answer first (featured snippet target), then numbered steps, each step = one action with "what goes wrong"
  • Comparison: Verdict first ("Choose A if... Choose B if..."), then detailed analysis
  • Listicle: Summary table above fold, consistent evaluation framework per item
  • Definition: "[Term] is..." in the first sentence, no preamble
  • Case study: Lead with the result number, then the story (PAS framework)
  • Pillar page: Table of contents, overview, then link to deep-dive articles
  • Service/landing page: PAS framework. Pain point first, agitate consequences, then solution
SEO Structure
  • Primary keyword in meta title, H1, first 100 words, 2-3 H2s. ~2% body density, naturally distributed.
  • Place a 40-60 word direct answer immediately after the most important H2 — targets featured snippets. "How" queries get ordered lists, "what is" gets paragraphs, comparisons get tables.
  • Weave 2-3 PAA questions into the article as H2/H3 headings with direct answers.
  • Include 3-5 internal links per 1,000 words. Use descriptive anchor text — never "click here."
  • Front-load value. The first screen is the highest-value real estate. No preamble paragraphs.
Long Article Strategy (1,500+ words)
  • Write section by section. Track what you've covered to prevent repetition and voice drift.
  • At least 30% of the article must contain details no generic AI could produce: the user's data, examples, opinions, experience.
Final Checks
  1. "So What?" test: For each major section — could anyone have written this, for anyone, about anything? If yes, inject specific knowledge.
  2. Self-check: Scan for blacklisted words, sections where every paragraph starts with a topic sentence, unnecessary section summaries, participial tack-ons. Fix before delivering.
Output

Clean markdown. Title + article content. Nothing else.

Language

Write in the language from the business context. If not specified, match the language of the user's messages.

Bundled references

Load from references/ only when the step or rule calls for them. Don't preload — each file is heavy enough to blow context if stacked.

Content type templates (references/content-types/) — load one after Phase 2:

  • Common: how-to.md, definition.md, comparison.md, listicle.md, pillar-page.md, faq-page.md, landing-page.md, service-page.md, case-study.md
  • Content and news: statistics-page.md, news-article.md, glossary-page.md
  • Commercial: alternatives-page.md, buying-guide.md, product-page.md, category-page.md, integration-page.md, location-page.md, programmatic-page.md
  • references/content-types-overview.md for the decision table across all 23 content types (load this FIRST if unsure which type to pick)

Writing technique modules (references/) — load when the matching Phase 4 rule needs more depth:

  • anti-slop-ruleset.md — full tiered banned vocab + structural tell list (when the inline anti-slop block isn't catching something)
  • voice-injection-playbook.md — voice and register techniques (when the draft reads flat)
  • information-gain-writing.md — the 30% rule and how to satisfy it (for the "So What" test)
  • serp-driven-writing.md — how Phase 1 research shapes the article (if the draft drifts from the SERP intent)
  • intent-matching.md — length and format decisions from intent (when the SERP is mixed)
  • eeat-signal-embedding.md — how to surface experience without a bio section
  • structured-data-snippets.md — JSON-LD and featured snippet formatting per content type
  • geo-optimization.md — optimizing for AI Overview / generative engine citation
  • navboost-engagement.md — engagement signal writing (dwell time, scroll depth)
  • quality-scoring.md — self-scoring rubric to run before delivery
  • writing-pipeline.md — the research → draft → edit loop
  • seo-optimization-layer.md — keyword placement, internal linking, metadata pass
  • fact-checking.md — how to verify every specific number and claim
  • human-input-framework.md — the knowledge-extraction questions (reinforces Phase 3)

© inhouseseo, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 34 other files (references) in skills/write-content of inhouseseo/superseo-skills.

  • SKILL.md
  • references/anti-slop-ruleset.md
  • references/content-types-overview.md
  • references/content-types/alternatives-page.md
  • references/content-types/buying-guide.md
  • references/content-types/case-study.md
  • references/content-types/category-page.md
  • references/content-types/comparison.md
  • references/content-types/definition.md
  • references/content-types/faq-page.md
  • references/content-types/glossary-page.md
  • references/content-types/how-to.md
  • references/content-types/integration-page.md
  • references/content-types/landing-page.md
  • references/content-types/listicle.md
  • references/content-types/location-page.md
  • references/content-types/news-article.md
  • references/content-types/pillar-page.md
  • references/content-types/product-page.md
  • … and 16 more

Open the folder on GitHubat commit 9cf22cc

Compare with similar skills

Write Content next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Write Content compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Write Content this skillinhouseseo/superseo-skills350—~3.1kAutomated safety check: PassApache-2.0
AI Content Creatorhuifer/claude-code-seo110—~1.7kAutomated safety check: NotesMIT
Blog AnalyzeAgriciDaniel/claude-blog2.3k1 repos~3.5kAutomated safety check: PassMIT
Blog CannibalizationAgriciDaniel/claude-blog2.3k1 repos~2.2kAutomated safety check: PassMIT
Blog WriteAgriciDaniel/claude-blog2.3k1 repos~5.6kAutomated safety check: PassMIT
BlogAgriciDaniel/claude-blog2.3k—~6.2kAutomated safety check: PassMIT

Similar skills

  • AI Content Creator

    huifer/claude-code-seo

    AI 内容创作专家,生成高质量的 SEO 和 GEO 优化内容,支持多种内容类型和风格. An agent skill from huifer/claude-code-seo.

    110 GitHub stars~1.7k tokensUpdated 9 mo ago
    Writing & ContentAuto-check: notes
  • Blog Analyze

    AgriciDaniel/claude-blog

    Audit and score blog posts on a 5-category 100-point scoring system covering content quality, SEO optimization, E-E-A-T signals, technical elements, and AI citation readiness.

    2.3k GitHub starsUsed in 1 repo~3.5k tokens
    Writing & ContentAuto-check passed
  • Blog Cannibalization

    AgriciDaniel/claude-blog

    Detect keyword cannibalization across blog posts by extracting primary keywords from titles and headings, clustering semantically similar targets, and flagging posts competing for the same search…

    2.3k GitHub starsUsed in 1 repo~2.2k tokens
    Writing & ContentAuto-check passed
  • Blog Write

    AgriciDaniel/claude-blog

    Write new blog articles from scratch optimized for Google rankings and AI citations.

    2.3k GitHub starsUsed in 1 repo~5.6k tokens
    Writing & ContentAuto-check passed
  • Blog

    AgriciDaniel/claude-blog

    Full-lifecycle blog engine with 31 sub-skills, 12 templates, 100-point scoring, and 5 agents.

    2.3k GitHub stars~6.2k tokensUpdated yesterday
    Writing & ContentAuto-check passed
  • Blog Brief

    AgriciDaniel/claude-blog

    Generate detailed content briefs for blog posts with target keywords, content outlines, competitive analysis, recommended statistics, image and chart suggestions, word count targets, internal…

    2.3k GitHub stars~3.2k tokensUpdated yesterday
    Writing & ContentAuto-check passed

More from inhouseseo/superseo-skills

All 11 skills in this repo
  • Content Brief

    inhouseseo/superseo-skills

    A skill your agent uses when planning a new article. An agent skill from inhouseseo/superseo-skills.

    350 GitHub stars~1.5k tokensUpdated 1 mo ago
    Auto-check passed
  • Eeat Audit

    inhouseseo/superseo-skills

    A skill your agent uses when auditing a page for E-E-A-T signals.

    350 GitHub stars~1.8k tokensUpdated 1 mo ago
    Auto-check passed
  • Expert Interview

    inhouseseo/superseo-skills

    A skill your agent uses when extracting first-party expertise from a subject-matter expert before writing content.

    350 GitHub stars~1.1k tokensUpdated 1 mo ago
    Auto-check passed
  • Featured Snippet Optimizer

    inhouseseo/superseo-skills

    A skill your agent uses when you want to win a featured snippet for a keyword you already rank for.

    350 GitHub stars~2k tokensUpdated 1 mo ago
    Auto-check passed
  • Improve Content

    inhouseseo/superseo-skills

    A skill your agent uses when rewriting or refreshing an existing page that's underperforming.

    350 GitHub stars~1.5k tokensUpdated 1 mo ago
    Auto-check passed
  • Keyword Deep Dive

    inhouseseo/superseo-skills

    A skill your agent uses when planning to rank for a specific keyword.

    350 GitHub stars~1.7k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Write Content

What does Write Content do?

A skill your agent uses when writing a complete SEO article. Write Content is an agent skill from inhouseseo/superseo-skills. Use when writing a complete SEO article.

When should I use Write Content?

Write Content fits situations like: writing a complete SEO article; tasks that involve Blog and article writing.

How do I install Write Content in Claude Code?

Run `npx skills add inhouseseo/superseo-skills --skill write-content -a claude-code`. Or copy the skill folder (skills/write-content in inhouseseo/superseo-skills) into .claude/skills/write-content in your project. Claude Code loads it when a task matches its description.

How do I install Write Content in Codex?

Run `npx skills add inhouseseo/superseo-skills --skill write-content -a codex`. Or copy the skill folder (skills/write-content in inhouseseo/superseo-skills) into .agents/skills/write-content in your project. Codex loads it when a task matches its description.

Can I use Write Content in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add inhouseseo/superseo-skills --skill write-content -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/write-content, .gemini/skills/write-content, .github/skills/write-content and .opencode/skills/write-content in your project.

What does Write Content need to run?

SKILL.md names no scripts, command-line tools or credentials: Write Content is instructions for the agent only.

Does Write Content access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Write Content safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Write Content use?

Write Content is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Write Content use?

About 3.1k tokens (SKILL.md is roughly 13k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 57k tokens, read only when the agent opens those files.

What are the alternatives to Write Content?

Skills that share tags, products or a category with Write Content: AI Content Creator (huifer/claude-code-seo, 110 stars), Blog Analyze (AgriciDaniel/claude-blog, 2.3k stars), Blog Cannibalization (AgriciDaniel/claude-blog, 2.3k stars) and Blog Write (AgriciDaniel/claude-blog, 2.3k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Write Content?

inhouseseo (a GitHub user) maintains it in inhouseseo/superseo-skills, which has 350 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on September 3, 2026.

Source: inhouseseo/superseo-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.