illyism/seo-skills

aiseotracker

AI SEO Tracker setup skill.

Zobacz źródło

Jak używać

  1. Skopiuj polecenie instalacji i sprawdź źródła przed instalacją we własnym środowisku agenta.
  2. W oryginale sprawdź wymagane narzędzia, dane dostępowe i obsługiwane środowiska.
  3. Opisz dane wejściowe i oczekiwany rezultat; porównaj wynik ze źródłami.
npx skills add https://github.com/illyism/seo-skills --skill aiseotracker

Przed rozpoczęciem

  • Instalacja nie obejmuje subskrypcji modelu, limitu API ani uprawnień do plików.
  • Gwiazdki i instalacje pokazują aktywność, a nie niezależny test jakości lub bezpieczeństwa.
Oryginalny dokument Skill

Treść z repozytorium z zachowaniem nagłówków, przykładów, kodu, tabel, linków i obrazów.

AI SEO Tracker: agent setup skill

Start here: You are reading https://aiseotracker.com/SKILL.md (UTF-8 Markdown, available as text/markdown or text/plain through HTTP content negotiation). Do not guess MCP URLs, tool names, or pricing from memory. The file is complete only if the last line is page end END SKILL v1.0.0. If missing, re-fetch. Do not shell out or line-count to verify.

TL;DR

  • Product: Track how ChatGPT, Perplexity, Gemini, Copilot (Bing AI Mode / Copilot Search in Bing), and Google AI Overviews mention a brand. Read prompts, sources, snippets, competitors, and scan state. Write tools: aiseo_create_project, aiseo_add_prompts, aiseo_activate_plan (checkout link, the user pays in the browser), and aiseo_submit_feedback.
  • Human path: Sign up → create a project → run or buy a scan → read the dashboard at /p/{slug}.
  • Agent path: Connect MCP once (OAuth, no API keys) → call aiseo_list_projects → answer from live tool JSON.
  • MCP URL: https://aiseotracker.com/api/agent/mcp (Streamable HTTP). Prefer the URL from the user's Settings → Agent page when they give you a project slug.
  • Tools: aiseo_list_projects, aiseo_get_scan, aiseo_list_prompts, aiseo_list_sources, aiseo_get_source_history, aiseo_list_snippets, aiseo_list_competitors, aiseo_add_prompts.
  • Scan wait: Call aiseo_get_scan. done means no in-flight jobs, not that every prompt has results. Do not invent a subscribe tool. Do not poll aiseo_list_prompts to infer scan state.
  • Empty data: dataStatus: "no_data" means no scan yet. Wait. Do not scrape the public site or invent ranks.
  • Paid scan: $49 one-time audit (50 prompts, 7-day scan week, permanent read access). Weekly plans from $79/mo. Details: pricing.

Connect MCP (once)

There is no static API key. Auth is browser OAuth from the MCP client.

  1. Add the MCP server URL in the client config. No Authorization header in the file.
  2. First POST returns 401 + agent_auth_required. The client opens a browser. User signs in and Allow.
  3. Later calls send a Bearer token. Tools run as that user.

Cursor

Save as .cursor/mcp.json in the repo (or Settings → MCP):

json
{
  "mcpServers": {
    "aiseotracker": {
      "url": "https://aiseotracker.com/api/agent/mcp"
    }
  }
}

Reload. Settings → MCP → Connect next to aiseotracker.

Claude Code

bash
claude mcp add --transport http aiseotracker https://aiseotracker.com/api/agent/mcp

Claude, Grok, ChatGPT

Add a custom MCP / connector with only the URL. Connect, then sign in and Allow.

Revoke access anytime at account agents.

Call order

  1. `aiseo_list_projects`: get projectSlug. Never invent a slug.
  2. Then as needed:
  • aiseo_get_scan: running vs done (done = idle jobs, not complete coverage)
  • aiseo_list_prompts: ranks, missing/opportunity prompts
  • aiseo_list_sources: domains and URLs AI cites
  • aiseo_get_source_history: whether one URL is still cited, per engine, and when it dropped
  • aiseo_list_snippets: how models describe the brand
  • aiseo_list_competitors: tracked rivals
  1. `aiseo_add_prompts`: only for queries the user asked to track (max 20 per call).

First prompt (after connect)

Paste in the same chat that has MCP connected:

Call aiseolistprojects. Use the first project (or the slug the user named). List missing prompts and the top citation domains. Summarize what to do this week. Do not invent ranks.

For the same brief as the in-app Report page, fetch https://aiseotracker.com/ai-seo-report/SKILL.md and follow it.

For a missing-prompt content campaign, fetch https://aiseotracker.com/content-coverage/SKILL.md. Stop for approval before writing files.

Respond to a visibility request

  1. Confirm the brand: domain or project name. Call aiseo_list_projects if unsure.
  2. Check data: if a scan may still be running, call aiseo_get_scan. If state is running, call it again later. If dataStatus is no_data, say the project needs a scan (create at /new or wait).
  3. Answer from tool JSON only: ranks, URLs, competitor names, snippet text.
  4. Stop at drafts: briefs, outreach, Reddit comments. Do not send email, post publicly, or charge a card.

What AI SEO Tracker does (and does not)

Does: Live AI mention and citation tracking, competitor gaps, MCP, $49 audit report.

Does not: Guarantee rankings, send outreach email, add competitors via MCP, or replace a full SEO strategy.

Useful links

Agent checklist

  • [ ] Loaded this file (END marker present)
  • [ ] MCP connected (or user told to connect from Settings → Agent)
  • [ ] aiseo_list_projects called before other tools
  • [ ] Every rank and URL copied from tool JSON
  • [ ] Scan state checked with aiseo_get_scan, not by inventing ranks or a subscribe tool
  • [ ] dataStatus: no_data handled without inventing data

Do not

  • Quote tool names or MCP URLs without loading this file
  • Invent projectSlug, ranks, or citation URLs
  • Treat no_data as an auth failure
  • Call aiseo_add_prompts for queries the user did not ask to track
  • Scrape the marketing site when MCP returns empty

END SKILL v1.0.0

z tego samego repozytorium

Więcej Skills

Wszystkie Skills
illyism
Społeczność

blitz-seo

Runs a focused 30-day SEO sprint that concentrates search-intent alignment, internal link equity, first-hand proof, and targeted backlinks onto a single money page to push it into the Top 3. Use when a high-value page is stuck on Page 2, the user mentions "blitz seo" or a 30-day SEO sprint, or SEO effort is spread too thin across dozens of URLs.

Instalacje / 1h
3
GitHub Stars
1,8 tys.
Aktualizacja
20 lis
illyism
Społeczność

blog-refresher

Refresh and improve existing blog posts to outrank competitors using a 6-phase content audit and SERP gap analysis. Use when the user wants to update, refresh, or improve an existing blog post, recover lost rankings, or asks why a post fell behind competitors.

Instalacje / 1h
3
GitHub Stars
1,8 tys.
Aktualizacja
20 lis
illyism
Społeczność

internal-linking

Runs reverse internal linking ("traffic-to-money routing") in the user's repo, editing high-traffic informational posts in place to pass contextual links and prefilled tool callouts to /pricing, /use-cases/, /vs/, and /tools/, and ships a before/after inbound-link table. Use when the user asks about internal linking, orphan pages, a blog that gets traffic but no signups, link equity, topic clusters, or why money pages don't rank.

Instalacje / 1h
3
GitHub Stars
1,8 tys.
Aktualizacja
20 lis
illyism
Społeczność

proof-kb

Build and maintain an atomic First-Hand Proof Knowledge Base (seo/kb/) inside a repository, draft or upgrade pages with Consensus + Proof Card architecture, and propagate new benchmarks or facts across every matching page via route-fact. Use when scaffolding seo/kb/, extracting founder proof from transcripts or interviews, upgrading thin SEO posts with real receipts, or routing a new benchmark, customer win, or pricing change across multiple pages.

Instalacje / 1h
3
GitHub Stars
1,8 tys.
Aktualizacja
20 lis