Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Jasper0122/agent-seo-kitnpx agentmods add skills/jasper0122/agent-seo-kit/geo-monitorWrote 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/jasper0122/agent-seo-kit/geo-monitor)<a href="https://agentmods.dev/skills/jasper0122/agent-seo-kit/geo-monitor"><img src="https://agentmods.dev/badge/skills/jasper0122/agent-seo-kit/geo-monitor/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/jasper0122/agent-seo-kit/geo-monitor"><img src="https://agentmods.dev/badge/skills/jasper0122/agent-seo-kit/geo-monitor.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 | $0.00159 | $0.02320 |
| Opus 5.5 | $0.00064 | $0.00928 |
| Sonnet 5.5 | $0.00032 | $0.00464 |
| Haiku 4.5 | $0.00016 | $0.00232 |
Grade A, and why
geo-monitor 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 1mo 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 — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GEO monitor
Sibling to seo-intake, same repo, same demand gate.
The reason this pipeline exists: losing in an AI answer is invisible. In search, a lost customer still leaves an impression with no click. In an answer they leave nothing at all: no click, no referrer, no log line. The only way to know it happened is to ask the questions yourself and write down what came back.
The cycle
node scripts/run-geo.mjs --dry-run # see the questions, spend nothing
node scripts/run-geo.mjs # registry -> engine -> answers + detection
node scripts/geo-dictionary.mjs --min 2 # harvest who was actually named (DRAFT only)
node scripts/redetect.mjs # re-read stored answers with a new dictionary, free
node scripts/geo-opportunities.mjs # three gap types, counted separately
node scripts/geo-ledger.mjs # merge into the persistent, idempotent ledger
run-geo takes --limit N and --cohort <name>. It is single threaded with a
gap between calls and a backoff on failure, because a throttled row and a
genuine absence look identical once they are written down, and one of them is
a lie. Failures are recorded as errors and excluded from every denominator.
The three signals, and which one you can actually move
| Signal | How it is read | How fast content moves it |
|---|---|---|
mentioned |
your brand name in the answer text | slow, depends on what the model knows |
recommended |
heuristic, see below | slower still |
cited |
your domain in the answer's sources | fastest: this is what a good page changes |
cited is the GEO equivalent of a backlink and the one thing publishing
directly affects. It is also why a page that renders empty to a crawler hurts
more here than in search: an unreadable page can never enter a citation list at
all.
Never mix the cohorts
Brand defence questions name the brand almost by construction. Averaged with generic questions they produce a healthy looking number that describes nothing that happened.
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.
- 1mo ago First seen · 206 lines · 159 tokens per session scan A a72fd4a7a08e
geo-monitor is a skill published in the GitHub repository Jasper0122/agent-seo-kit (8 stars, last pushed 1mo ago), licensed MIT. It adds 159 tokens to every session and 2,320 once invoked, about $0.0006 per session on Opus 5.5. 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
fire-your-seo-agency
A procedure for improving how a website appears in search engines and how AI answer systems find and cite it. It covers search, answer-engine, generative-AI, and Naver visibility.
optimise-seo
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.
seo-aeo
Audit supplied pages, documents, or content sets for direct-answer completeness, question and intent coverage, clarity, extractability, claim support, and visible-content consistency against a pinned AI-search Research Pack. Use for AEO, answer-engine optimization, answer readiness, answer extraction, featured-answer…
seo-geo
Audit supplied pages, entity records, source captures, and citation observations for generative-engine entity consistency, evidence traceability, citation suitability, cited-source alignment, and documented engine controls against a pinned AI-search Research Pack. Use for GEO, generative-engine optimization, AI…
seo-plan
Use when sequencing validated SEO findings and approved action-plan items into an owner-based roadmap with dependencies, review gates, measurable outcomes, and risk controls; do not use it to invent work, fixed uplift forecasts, or a one-size-fits-all calendar.
seo-audit
A checklist-based SEO review for a website. SEO, or search engine optimization, is the work of improving a site so search engines can understand and rank it.