optimise-seo
Skill6 starsChanged 3 months agoApache-2.0View on GitHub
Use when explicitly authorized to implement a bounded, approved SEO action in a codebase or content system, verify the exact change, and report its limitations; do not use for broad audits, self-approved production changes, or outcome guarantees.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- HK-hub/AgentSkillstheir own version, first seen Mar 2, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- HK-hub/AgentSkills★ 6same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository27 other files it gives its agents.
- llms.txtllms.txt
- ai-search-researchSkill
- ai-visibility-monitorSkill
- seo-action-planSkill
- seo-aeoSkill
- seo-agenticSkill
- seo-architectureSkill
- seo-auditSkill
- seo-authoritySkill
- seo-commerceSkill
- seo-competitor-pagesSkill
- seo-contentSkill
- seo-geoSkill
- seo-hreflangSkill
- seo-imagesSkill
- seo-localSkill
- seo-news-discoverSkill
- seo-pageSkill
- seo-performanceSkill
- seo-planSkill
- seo-programmaticSkill
- seo-researchSkill
- seo-schemaSkill
- seo-sitemapSkill
- seoSkill
- seo-technicalSkill
- seo-videoSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
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
- Very specific: line 10
Read markdown-delivery-contract.md for every task.
- Very specific: line 19
Read references/implementation-gate.md before modifying a workspace.
- Distinctive: line 20
Read verification-modes.md before choosing per-action, batch-final, or production-comparison verification.
- Distinctive: line 21
Read external-closeout.md before identifying or releasing a production candidate, especially from a dirty or generated worktree.
- Distinctive: line 34
If any is missing, prepare an implementation plan or patch proposal only.
- Distinctive: line 49
Never promise crawl, index, rank, citation, traffic, revenue, or conversion results.
- Distinctive: line 50
Do not expand the task from one approved action into a broad “fix everything” change.
- Distinctive: line 51
Do not overwrite unrelated user work, secrets, configurations, or analytics data.
- Distinctive: line 52
Do not replace a dirty worktree with a clean copy or edit a generated artifact without identifying its source/generator ownership.
- Very specific: line 53
Treat crawler policies, canonicals, robots/noindex, redirects, privacy/security, pricing, health/legal/financial claims, and localization …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
SEO Implementation Coordinator
Markdown delivery contract
Read markdown-delivery-contract.md for every task. Write a substantive .md document for every material work step and register its direct destination in the work index.md. For public-site work, apply the owner-required Markdown discovery profile across the full eligible inventory; preserve native standards and immutable evidence formats.
Lane outputs: implementation.md; local-acceptance.md; release-verification.md (combine small related steps with stable section links). Implement the approved Markdown corpus plus conventional SEO changes to the exact requested scope, maintaining source parity, indexing policy, generated maps and rollback. Validate local projection and all affected live resources after authorized release. No completion from files alone.
Purpose
Apply or prepare a specific, user-authorized SEO change after its evidence, product decision, owner, scope, acceptance criteria, verification, and rollback are known. This skill coordinates the chain from an approved action to local acceptance and, only when separately authorized, exact-candidate release and live closeout. It is not an audit, an auto-publisher, or implied deployment authority.
Read references/implementation-gate.md before modifying a workspace.
Read verification-modes.md before choosing per-action, batch-final, or production-comparison verification.
Read external-closeout.md before identifying or releasing a production candidate, especially from a dirty or generated worktree.
Authorization gate
Require all of the following before changing files or systems:
- explicit user/authorized-owner request to implement the named action;
- validated upstream finding/action plan or equivalent evidence record;
- exact target repository/content system and permitted scope;
- owner, acceptance criteria, verification plan, risk, and rollback;
- facts/copy/schema/data approved for the target locale.
- independent authorization states for implementation, release, and provider mutation.
If any is missing, prepare an implementation plan or patch proposal only. Do not infer approval from an audit result, a pending action-plan artifact, a backlog item, or the desire to “improve SEO.”
Implementation workflow
- Reconfirm scope. Quote the action ID/objective, target files/URLs, intended behavior, non-goals, and success/guardrail conditions. Inspect the current state; preserve unrelated work.
- Choose the owning lane. Content edits follow
seo-content; technical changes followseo-technical; markup followsseo-schema; locale annotations followseo-hreflang; sitemap changes followseo-sitemap; product/feed, local, video, news/Discover, agentic task-flow, architecture, and authority work follows its matching specialist. Measurement-only actions followseo-performanceorai-visibility-monitorand never become implementation by implication. Do not silently substitute a different change. - Make the smallest reversible change that satisfies the decision. Modify only the approved files/data. “Smallest” cannot shrink or replace an explicit product-owner end state. Preserve facts, locale, consent/accessibility, analytics, and security behavior. Do not add
llms.txt, crawler directives, schema, pages, links, or copy merely because they sound SEO-related. For an approved broad rank-higher/site optimization, “complete” also means the owned baseline surfaces were handled: query-to-page ownership, unique title, semantically complete meta description, canonical/indexability, and hostname favicon/site identity. Implement them from approved facts and owned brand assets or record the exact blocking dependency; do not silently omit them as minor details. When the approved title, description, or H1 is exact, bind that copy into the live release plan instead of accepting presence-only validation. - Preserve source and candidate identity. In a dirty worktree, identify authored sources, generators, and generated outputs before editing. Never reset, clean, overwrite, or revert unrelated work. Record a candidate manifest containing expected source paths, generated outputs, build identity/hash, corpus summary, and
gitDirty=1when applicable; compare post-build generated diffs with the initial worktree. - Verify the implementation. Choose a verification mode and run the complete declared gate. Batch-final verification accumulates every action's criteria and reruns the entire final gate after a failure. Match the verification universe to the change universe; record every attempt and what remained unverified.
- Close externally only when separately authorized. Discover the actual deployment mechanism, verify the candidate identity, deploy exactly that candidate, record the deployment ID and alias, and run
seo-technicallive verification over the affected universe. Provider operations are a further authorization phase owned byseo-performanceand require their own receipt. - Prepare handoff. Report changed files, candidate identity, exact behavior, test attempts, release/live evidence when authorized, provider-operation evidence when authorized, pending outcomes, rollback, and comparison prerequisites. If a later phase lacks authorization, name that exact closeout blocker; when authorization arrives, resume this same chain.
Guardrails
- Never promise crawl, index, rank, citation, traffic, revenue, or conversion results.
- Do not expand the task from one approved action into a broad “fix everything” change.
- Do not overwrite unrelated user work, secrets, configurations, or analytics data.
- Do not replace a dirty worktree with a clean copy or edit a generated artifact without identifying its source/generator ownership.
- Treat crawler policies, canonicals, robots/noindex, redirects, privacy/security, pricing, health/legal/financial claims, and localization as high-risk changes needing explicit review and rollback.
Output
Return an implementation receipt: decision/action/evidence reference, files/systems changed, candidate manifest, acceptance verification and attempts, external/production checks, release/live receipt when present, provider receipt when present, pending outcomes, known limitations, and rollback. If no implementation was authorized, return a reviewable plan rather than making changes.
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
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.

