agentleFS
lsidorelsidore/seo-md/skills/seo-programmatic/SKILL.md

seo-programmatic

Skill1 starsChanged 4 months agoMITView on GitHub

Use when the backlog's top opportunity is a programmatic or glossary cluster (comparisons, alternatives, glossary terms, for-industry pages), typically invoked by seo-run. Ships one capped batch; refuses to mass-produce thin pages.

32 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use seo-image-gen, seo-agentic and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • timruebenachSkill

    seo-image-gen

    AI image generation for SEO assets: OG/social preview images, blog hero images, schema images, product photography, infographics. Powered by Gemini via nanobanana-mcp. Requires banana extension installed. Use when user says \"generate image\", \"OG image\", \"social preview\", \"hero image\", \"blog image\", \"product photo\", \"infographic\", \"seo image\", \"create visual\", \"image-gen\", \"favicon\", \"schema image\", \"pinterest pin\", \"generate visual\", \"banner\", or \"thumbnail\".

    Also in 12 projects that use this

  • AgriciDanielSkill

    seo-agentic

    Audit and fix agent readiness: the Lighthouse Agentic Browsing fraction, accessibility tree for agents, robots.txt and Content-Signal for AI agents, WAF treatment of agent traffic, llms.txt, Markdown delivery, ai-catalog.json, /.well-known discovery files, and WebMCP tools. Use when the user asks whether AI agents can use or browse the site, or about llms.txt or WebMCP. Exclude AI citability and brand signals (seo-geo) and commerce protocol depth (seo-ecommerce).

    Also in 4 projects that use this

  • timruebenachSkill

    seo-flow

    FLOW framework integration — evidence-led SEO using the Find → Leverage → Optimize → Win loop. Surfaces stage-specific AI prompts from the FLOW knowledge base (41 prompts, CC BY 4.0). Use when user says "FLOW", "FLOW framework", "seo flow", "evidence-led SEO", "find leverage optimize win", or wants stage-specific SEO prompts.

    Also in 10 projects that use this

  • mangollcSkill

    seo-backlinks

    Backlink analysis and link building strategy: competitor backlink profiling, outreach plan generation, monthly SEO reporting framework, toxic link identification, link opportunity scoring, and unlinked mention discovery.

    Also in 13 projects that use this

  • indranilbanerjeeSkill

    seo-drift

    Compare two SEO exports (Search Console, two dates): gainers, losers, drift classes. \"what moved after the core update\"

    Also in 11 projects that use this

  • mangollcSkill

    seo-local

    Local SEO optimization: Google Business Profile post generation, 30-day posting calendars, NAP consistency checks, local citation audits, review response templates, local keywords.

    Also in 14 projects that use this

  • +6 more from timruebenach

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 30 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No verified search volume for the individual pages (a glossary term nobody searches is dead weight that dilutes the whole site).

    line 13
  2. Very specific:

    If blocked by these, say so, mark the topic skipped with the reason in angle, and suggest what would unblock it.

    line 22
  3. Very specific:

    Read SEO.md at the project root, in full.

    line 27
  4. Distinctive:

    Never invent product or audience facts that are not in SEO.md.

    line 30
  5. Very specific:

    Default paths (overridable in SEO.md § State): backlog .seo/backlog.json.

    line 31
  6. Distinctive:

    If none: report and suggest seo-plan.

    line 36
  7. Distinctive:

    Create a table (one row per future page) with the columns that will make each page unique: facts, numbers, features, pricing, quirks, use …

    line 47

“Common” is measured: kept by at least 6 in 10 of the 30 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

seo-programmatic: ship one capped batch of programmatic pages

When NOT to use this (read first)

Programmatic SEO is the fastest way to get a site deindexed when done wrong. Do NOT proceed if ANY of these holds:

  • No verified search volume for the individual pages (a glossary term nobody searches is dead weight that dilutes the whole site).
  • Nothing UNIQUE to say per page: if two pages would share 80% of their content, the cluster is one article, not a page set.
  • The site is young: fewer than ~10 indexed editorial pieces. Programmatic pages ride on existing site trust; they do not create it.
  • The goal is page count. Google's spam policies name "scaled content abuse" explicitly; mass thin pages put the entire domain at risk, not just the batch.

If blocked by these, say so, mark the topic skipped with the reason in angle, and suggest what would unblock it.

Before anything

  1. Read SEO.md at the project root, in full. It is the only source of truth for product, audience, voice, rules, and paths. If it does not exist: stop and tell the user to run the seo-init skill (or fill SEO.template.md by hand).
  2. Never invent product or audience facts that are not in SEO.md.
  3. Default paths (overridable in SEO.md § State): backlog .seo/backlog.json.

Step 1: Pick the cluster and pattern

Take the top-priority todo of type programmatic or glossary from the backlog. If none: report and suggest seo-plan. STOP.

PatternOnly when
Glossaryeach term has verified definition-intent volume
X vs Y / alternativesthe ICP genuinely weighs these options; you can be honest about where the competitor wins
For-[industry]each industry has distinct, nameable needs (not the same page with a swapped noun)
Templates / exampleseach template is genuinely usable, not a placeholder

Step 2: Build the dataset BEFORE any page

Create a table (one row per future page) with the columns that will make each page unique: facts, numbers, features, pricing, quirks, use cases. Research each row for real (WebSearch, product pages, real screenshots per the SEO.md screenshot rule).

The row rule: a row you cannot fill with unique, verified values is a page that does not get created. This single rule is what separates programmatic SEO from spam.

Step 3: Generate template + pages

  • One template in the host content system (SEO.md § Content System), pages generated from the dataset rows.
  • Per-page minimum UNIQUE content (shared boilerplate does not count toward it): glossary 400 words; comparisons and alternatives 700 words; for-industry 700 words.
  • Each page: meta title 50-60 chars with its keyword, description 140-160, voice and banned patterns per SEO.md.
  • A hub page links every child; siblings cross-link where relevant; 1-2 published articles link to the hub.

Step 4: Batch discipline

  • Cap per run: SEO.md § Content Mix & Cadence (default 10 pages). A 40-term glossary ships in 4+ runs, weeks apart.
  • One batch = one coherent cluster. Never mix patterns in a run.
  • Stagger publishedAt across the batch (realistic past window, at most one page dated today), exactly as in seo-write's batch rule.

Step 5: Quality gate (blocking)

  • Pick 3 random pages from the batch; verify each against the unique-content minimum (sed + wc technique from seo-write, minus the shared boilerplate).
  • Banned-pattern grep clean across the batch.
  • Diff two sibling pages: if the diff is mostly the keyword swapped, the batch fails.
  • Project build passes.

Then mark the backlog entry done (or update it with remaining rows if the cluster continues) and publish per the SEO.md publish mode.

Red flags: stop and re-read the top section

  • "The volume is probably there for most terms"
  • "I can raise the cap just this once, the pages are ready anyway"
  • "The boilerplate is good enough content"
  • "Swapping the industry name is basically customization"

Never do

  • Uncapped generation, whatever the excuse.
  • Pages for zero-volume or unverified-volume terms.
  • Near-duplicate siblings.
  • Publishing the whole batch dated the same day.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.