Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add OptimizerTeam/agent-skills --skill ai-presence-auditorgit clone --depth 1 https://github.com/OptimizerTeam/agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/optimizerteam/agent-skills/ai-presence-auditor)<a href="https://agentmods.dev/skills/optimizerteam/agent-skills/ai-presence-auditor"><img src="https://agentmods.dev/badge/skills/optimizerteam/agent-skills/ai-presence-auditor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/optimizerteam/agent-skills/ai-presence-auditor"><img src="https://agentmods.dev/badge/skills/optimizerteam/agent-skills/ai-presence-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00104 | $0.01112 |
| Opus 5 | $0.00052 | $0.00556 |
| Sonnet 5 | $0.00021 | $0.00222 |
| Haiku 4.5 | $0.00010 | $0.00111 |
Grade A, and why
ai-presence-auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Presence Auditor
Customers increasingly ask an AI assistant "who should I call for a plumber in Austin?" instead of scrolling Google. AI names only a handful of businesses — you're either in the answer or you're invisible. This skill audits where a local business stands across the surfaces AI actually reads, and returns a prioritized list of fixes.
What you need (inputs)
- Business name
- Primary service(s)
- City / service area
- Website URL
- 2–3 competitors (optional but useful)
How AI actually picks local businesses (the model the fixes follow from)
There is no single "AI" — different assistants read different sources, so "get found by AI" is a multi-surface job, not one listing:
- ChatGPT leans on the Bing index + your website + Yelp/Foursquare structured data — not primarily Google Business Profile.
- Google AI Overviews / AI Mode / Gemini lean on your Google Business Profile + the local pack.
- Microsoft Copilot ← the Bing ecosystem. Siri / Apple Maps ← Apple Business Connect.
A few ground rules that shape every recommendation:
- AI recommends a small set (~3–5) of businesses, and often not the Google #1. Being cited is not the same as being recommended.
- Reviews act as a trust filter — roughly 4+ stars is table stakes; thin or below-average review profiles get left out. (There's no proven exact numeric cutoff — don't invent one.)
- Multi-source consistency beats any single listing: consistent name/address/phone (NAP) across Google, Bing, Apple, Foursquare, Yelp, and directories is what lets the AI trust the entity.
- Reality check: AI-sourced recommendations are still a small share of local demand today versus the Google local pack and the phone. Treat this as a growing edge, not the whole game — and say so.
The audit
- Run representative buyer prompts. Across ChatGPT + Perplexity + Google AI (and Gemini/Copilot if available), use real customer phrasing: "best emergency plumber in {city}", "who should I call for {problem} in {city}", "top-rated {service} near {neighborhood}". For each prompt record: Is the business named? Which competitors are named? What reason or source does the AI give?
- Diagnose the surfaces. Check presence + completeness on Google Business Profile, Bing Places, Apple Business Connect, Foursquare, and Yelp. Flag any that are missing or unclaimed.
- Check NAP consistency (name, address, phone) across those registries — mismatches confuse the entity and suppress recommendations.
- Check reviews — volume, recency, average rating, and response rate. Below-average or thin → likely filtered out regardless of everything else.
- Check the website. Does it state, in plain text, exactly what you do and where ("24/7 emergency plumbing in {city}")? Are there service and service-area pages, and an FAQ that matches how people ask AI? Vague "quality solutions for all your needs" copy is effectively invisible to AI.
- Check third-party mentions/citations — is the business named in local "best-of" round-ups, directories, and review sites the AI pulls from?
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 82 lines · 104 tokens per session scan A effae6a5752a
ai-presence-auditor is a skill published in the GitHub repository OptimizerTeam/agent-skills (3 stars, last pushed 2mo ago), licensed MIT. It adds 104 tokens to every session and 1,112 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
amazing-seo-skill
SEO + AEO + GEO analysis for any website (SaaS, e-commerce, local, publisher, agency). Full-site audits, single-page deep-dive, technical SEO (crawl, index, CWV with INP, JS-rendering), Schema.org detection/validation/gen, E-E-A-T content quality, image optimization, sitemap, hreflang, programmatic SEO, competitor…
onsite-audit
Use to audit on-page SEO for a URL or a whole site section - "audit this page", "on-page check", /organic-os:onsite-audit. Read-only; needs no credentials for public checks.
hoo-task-board
Use to view or mirror the work queue - "task board", "show the queue", /organic-os:task-board. Mirrors to Notion when that connector exists; local queue.md is the source of truth.
ce-image
Use to create the featured image / social card for a drafted post - "make the featured image", /organic-os:image, or ce-produce step 4.
reset
Use when the user wants to remove a site from organic-os, deregister a site, uninstall, "reset organic-os", "remove my site", "start over", or runs /organic-os:reset. Guided teardown that shows what exists and, per user choice, deregisters the site from the local registry - it never deletes the brain repo or revokes…
ansvisor-aeo-coach-standalone
Standalone (no-MCP) version of the Ansvisor AEO Coach. Use this only when the user's Claude client cannot connect to the Ansvisor MCP server (e.g. claude.ai web without a Connector configured). Fetches live data from the Ansvisor REST API directly with the user's API key via code execution. For clients that support…