seosorted
Skill0 starsChanged 39 days agoMITView on GitHub
Entry point for SEOsorted.ai, the SEO and AI-search visibility platform. Use when the user mentions SEOsorted, wants to start or plan SEO for their site, asks "what should I do next for SEO", wants a status check of their workspace, or asks which SEOsorted skill to use. Connects the SEOsorted MCP server, reads the workspace, and routes to the right skill in the proper order (audit → plan → write → publish → measure → optimize).
In this repository36 other files it gives its agents.
- ai-citation-outreachSkill
- ai-crawler-analyticsSkill
- ai-visibility-trackerSkill
- backlink-outreachSkill
- backlinksSkill
- brand-voice-setupSkill
- bulk-content-runSkill
- cannibalization-checkSkill
- comparison-pagesSkill
- competitor-analysisSkill
- content-calendarSkill
- content-refreshSkill
- domain-authority-checkSkill
- find-questionsSkill
- keyword-researchSkill
- link-building-playbooksSkill
- link-reclamationSkill
- llms-txt-generatorSkill
- page-seo-auditSkill
- publish-to-cmsSkill
- seo-auditSkill
- seo-master-planSkill
- seo-reportSkill
- serp-analysisSkill
- site-settingsSkill
- statistics-pageSkill
- technical-audit-reportSkill
- technical-seo-checkSkill
- topic-clustersSkill
- update-and-republishSkill
- write-articleSkill
- youtube-channel-auditSkill
- youtube-citation-strategySkill
- youtube-creator-outreachSkill
- youtube-keyword-researchSkill
- youtube-outliersSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Call get_site_overview. If the tool is missing or returns an auth error:
- Distinctive: line 15
No account yet: send the user to https://seosorted.ai to create one.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
SEOsorted
SEOsorted.ai researches, writes, publishes and measures SEO content for a website, and tracks how often AI engines (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews) cite it. This plugin gives the agent that platform through the SEOsorted MCP server plus one skill per job.
1. Make sure the connection works
Call get_site_overview. If the tool is missing or returns an auth error:
- Claude Code: run
/mcp, select seosorted, and authenticate. A browser window opens on seosorted.ai; the user signs in and picks one workspace (one website). Every tool then works on that workspace. - No account yet: send the user to https://seosorted.ai to create one. Onboarding (website, business, competitors) is what makes everything downstream good — do not skip it.
- Headless/CI: a workspace API key (
sk_ss_live_…, created in Settings → API keys) can be sent asAuthorization: Bearer <key>.
2. Read the workspace before recommending anything
From get_site_overview, note: website and business, connections (CMS for publishing, Google Search Console, GA4), content counts by status, keywords by opportunity bucket, next scheduled posts, last audit, quota left this month, and any active generation run.
Flag gaps plainly: no Search Console = no performance data or striking-distance work; no CMS = articles cannot auto-publish; nothing scheduled = the publishing cadence has stopped.
3. Run the loop in this order
| Step | Skill | Why this order |
| 1 | seo-audit | Fix technical problems before adding pages to a broken site |
| 2 | keyword-research → topic-clusters | Pick topics by volume vs difficulty, grouped into clusters |
| 3 | cannibalization-check | Never write two pages Google treats as one query |
| 4 | write-article or bulk-content-run | Generate with the workspace's voice, images and internal links |
| 5 | content-calendar → publish-to-cms | Keep a steady cadence; nothing publishes unscheduled |
| 6 | seo-report | Measure with Search Console and GA4 |
| 7 | content-refresh | Improve pages on page 2 or decaying — fastest wins |
AI-search work runs alongside: ai-visibility-tracker, ai-citation-outreach, ai-crawler-analytics, youtube-citation-strategy.
Authority and positioning: competitor-analysis, domain-authority-check, serp-analysis, comparison-pages, backlinks, link-reclamation, backlink-outreach, link-building-playbooks, statistics-page.
Editing what exists: update-and-republish (text, meta, images, republish), page-seo-audit (any URL).
YouTube: youtube-keyword-research, youtube-outliers, youtube-channel-audit, youtube-creator-outreach.
Planning and setup: seo-master-plan (90-day plan), brand-voice-setup and site-settings (do these before bulk writing).
Free checks (work without an account, better with one): technical-seo-check, find-questions, llms-txt-generator.
4. Routing rules
- "Write about X" with one keyword →
write-article. Ten or more keywords →bulk-content-run. - "Why did traffic drop" / "what's working" →
seo-report, thencontent-refresh. - "Are we showing up in ChatGPT" →
ai-visibility-tracker; "get us mentioned" →ai-citation-outreach. - "Fix / edit / update this article" →
update-and-republish. "Check this URL" →page-seo-audit. - "What's the DR of…" →
domain-authority-check. "Who ranks for…" / "where do we rank" →serp-analysis. - Anything that publishes, schedules, deletes, or spends credits: state what will happen and the credits it uses, and get a yes first. Tools that change things outside SEOsorted require
confirm: true; set it only after the user agrees, and usedryRun: trueto preview.
Ground rules for every skill
- Numbers come from tools, never from memory. If a tool returns nothing, say so — an empty result is a finding.
- Quota is real money for the user. Quote the remaining count the tools return before large runs.
- YMYL topics (health, finance, legal): keep claims attributable to sources, never promise outcomes.
- Work happens in one workspace. If the user means a different site, they must reconnect and pick that workspace.
See references/tools.md for the full tool list.
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.