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 T4wroot/agentic-seo --skill descriptiongit clone --depth 1 https://github.com/T4wroot/agentic-seoWrote 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/t4wroot/agentic-seo/description)<a href="https://agentmods.dev/skills/t4wroot/agentic-seo/description"><img src="https://agentmods.dev/badge/skills/t4wroot/agentic-seo/description/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/t4wroot/agentic-seo/description"><img src="https://agentmods.dev/badge/skills/t4wroot/agentic-seo/description.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 | $0.00095 | $0.00889 |
| Opus 5 | $0.00048 | $0.00445 |
| Sonnet 5 | $0.00019 | $0.00178 |
| Haiku 4.5 | $0.00010 | $0.00089 |
Grade A, and why
meta-description 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.
This is a copy
100% identical to meta-description — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SEO On-Page: Meta Description
Guides optimization of the meta description tag for search engines and SERP display.
When invoking: On first use, if helpful, open with 1–2 sentences on what this skill covers and why it matters, then provide the main output. On subsequent use or when the user asks to skip, go directly to the main output.
Scope (On-Page SEO)
- Meta description: CTA; unique value; target keyword; unique per page
Length by Language
Google truncates by pixel width (~920px desktop, ~680px mobile), not character count. Character limits are approximate—CJK chars are wider (~2× Latin), so fewer fit in the same pixels.
| Script / Language | Meta description (chars) | Notes |
|---|---|---|
| Latin (English, Spanish, French, etc.) | 150–160 | Desktop ~158; mobile ~120 |
| CJK (Chinese, Japanese, Korean) | 75–100 | Full-width chars; 70–80 conservative; 90–100 on some locales/fonts; use pixel checker when available |
| Cyrillic (Russian, etc.) | 140–155 | Slightly wider than Latin |
| Arabic, Hebrew | 70–90 | RTL; variable width |
Pixel tools: Use a pixel-accurate meta tag checker for CJK—font and locale affect display; character counts vary by source (65–80 to 90–120 in practice).
Multilingual: Use locale-specific limits; localize, don't just translate. See localization-strategy, translation.
Initial Assessment
Check for project context first: If .claude/project-context.md or .cursor/project-context.md exists, read it for brand voice and target keywords.
Identify:
- Page type: Homepage, landing, blog, product, etc.
- Primary keyword: Target search query
- Language / script: Apply length rule above
- CTA: Primary action (sign up, learn more, buy, etc.)
Best Practices
| Item | Guideline |
|---|---|
| Length | Per language (see table above); ~150 chars sweet spot for Latin; truncates beyond pixel limit |
| Unique | One per page; no duplicate descriptions |
| Intent | Answer "why should I click?"; match search intent |
| CTA | Include clear call-to-action when relevant |
| Keyword | Naturally include target keyword |
| Content | Include author, date, price where relevant |
| Impact | Does not affect ranking; well-written descriptions improve CTR 5–10% |
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 · 73 lines · 95 tokens per session scan A bb19bbfaa2d8
meta-description is a skill published in the GitHub repository T4wroot/agentic-seo (15 stars, last pushed 7d ago), licensed MIT. It adds 95 tokens to every session and 889 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to meta-description, differing in 0 lines, and is treated as a copy.
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.
content-thicken
Evidence-driven blog driver. Takes ONE target from the fused search plus AI-answer content library, pulls the real questions it has to answer, drafts a thick long-form guide against a template contract, validates it deterministically, and stops at preview. Two modes, thicken an already-earning post IN PLACE (never…
backlink-outreach
Find, evaluate, pitch and track natural backlink and content partnerships. Prospects come from the GEO citation data rather than a generic blog search: the targets are the pages an AI already cites when answering your category questions. Research runs on the Monid tool layer (web search and scrape, authority metrics…
geo-monitor
Run and read the GEO pipeline, which measures whether AI answer engines mention, recommend and cite your site. Puts a fixed registry of real user questions to an answer engine through the Monid tool layer, detects the three signals plus competitors, and turns "a rival is named and we are not" into a tracked work…
seo-intake
Run and read the SEO intake pipeline. Pulls the organic keyword set for your domain and for each competitor through the Monid tool layer, computes the gap locally, and routes every keyword to the one action that can help it (write-new / striking / build-depth / defend / noise). Use when asked to refresh the SEO queue…
bing-duplicate-content-fix
Audit article groups, WordPress categories, or supplied URLs for copied content, near-duplicate templates, search-intent overlap, and keyword cannibalization; prepare exact SEO/AEO/GEO fixes; apply only approved WordPress changes; verify saved and public results; and request a Bing recrawl when authorized. Use for…