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.
git clone --depth 1 https://github.com/minosdevs/seo-geo-kitWrote 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/commands/minosdevs/seo-geo-kit/seo-blog)<a href="https://agentmods.dev/commands/minosdevs/seo-geo-kit/seo-blog"><img src="https://agentmods.dev/badge/commands/minosdevs/seo-geo-kit/seo-blog.svg" alt="Measured on agentmods" 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 | $0.00024 | $0.00699 |
| Opus 5 | $0.00012 | $0.00349 |
| Sonnet 5 | $0.00005 | $0.00140 |
| Haiku 4.5 | $0.00002 | $0.00070 |
Grade A, and why
seo-blog 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 7d 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.
What it actually says
Ecris un article de blog en mode CONTENU.
Sujet : $ARGUMENTS
Si vide, prends le premier trou d'intention du dernier rapport seo/AUDIT-*.md. S'il n'y a pas de
rapport, analyse le site rapidement pour identifier un trou reel, ou demande le sujet.
Utilise le skill seo-geo-audit (mode CONTENU) plus references/blog-playbook.md.
Deroule :
- Comprends le site avant d'ecrire. Lis la landing, 2 ou 3 articles existants, le CLAUDE.md s'il existe. Releve la voix, la longueur type, le format exact du front-matter, l'arborescence du contenu, et comment un article est cable au sitemap, au feed et au JSON-LD. Tu copies ce format, tu n'en inventes pas un nouveau.
- Verifie qu'aucun article existant ne couvre deja l'intention. Si oui, propose une mise a jour de l'existant plutot qu'un doublon qui cannibalise.
- Cherche pour de vrai avec
WebSearch: qui se positionne, quel angle, quelle profondeur, quelles questions connexes. Tu dois savoir ce que tu dois battre. Note les sources avec leur URL. - Ecris : une intention par article, la reponse dans les 80 premiers mots, H2 formules comme des questions, blocs extractibles, donnees en tableaux, exemple concret et chiffre, FAQ finale, conclusion actionnable, un seul CTA.
- Cable tout : front-matter complet,
BlogPosting+BreadcrumbList(+FAQPagesi la FAQ est visible), 3 a 5 liens internes contextuels, 2 a 4 liens externes vers les sources, ajout au sitemap, au feed et allms.txt, image OG,altsur les images. - Version bilingue si le site l'est : une adaptation, pas une traduction mot a mot, avec un
altSlugreciproque pour lehreflang. - Verifie : typecheck, build, l'URL repond 200, le JSON-LD est du JSON valide, tous les liens internes pointent vers des pages qui existent.
Non negociable :
- Zero hallucination. Chaque chiffre, date, prix ou nom cite vient d'une source que tu as lue ou du repo. Dans le doute, tu n'ecris pas le chiffre.
- Pas de remplissage, pas d'intro qui tourne autour du pot, pas d'adjectif sans preuve.
- Pas de tirets cadratins ni de doubles tirets dans la copie.
- Si tu n'as rien a apporter que le top 5 actuel n'a pas, dis-le et propose un autre sujet.
Termine par : le chemin des fichiers crees, la requete visee, et ce qu'il reste a faire cote humain (image, relecture, publication).
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.
- 7d ago First seen · 44 lines · 24 tokens per session scan A 3fd7004f488b
seo-blog is a command published in the GitHub repository minosdevs/seo-geo-kit (2 stars, last pushed 21d ago), licensed MIT. It adds 24 tokens to every session and 699 once invoked, about $0.0001 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 commands, from other repositories
geo
Full GEO optimization pipeline - analyze, rank, rewrite, and generate schema for any URL or content.
geo:optimize
Optimize a local content file for GEO without full audit.
rank
SEO/GEO/AEO audit, analysis, and optimization. The most comprehensive search toolkit for Claude Code.
rank-aeo
Answer engine optimization. Featured snippets, voice search, People Also Ask.
create-content
Generate SEO + AEO + GEO-optimized content for a topic or keyword.
create-topic
Research and generate a complete topic plan with keyword mapping, audience, angle, and competitive positioning.