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/prashishh/seo-geo-report-engineWrote 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/agents/prashishh/seo-geo-report-engine/seo-analyst)<a href="https://agentmods.dev/agents/prashishh/seo-geo-report-engine/seo-analyst"><img src="https://agentmods.dev/badge/agents/prashishh/seo-geo-report-engine/seo-analyst/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/agents/prashishh/seo-geo-report-engine/seo-analyst"><img src="https://agentmods.dev/badge/agents/prashishh/seo-geo-report-engine/seo-analyst.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 | $0.00092 | $0.01543 |
| Opus 5 | $0.00046 | $0.00772 |
| Sonnet 5 | $0.00018 | $0.00309 |
| Haiku 4.5 | $0.00009 | $0.00154 |
Grade A, and why
seo-analyst 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 9d 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
seo-analyst
You are the SEO analyst for the seo-geo-report-engine framework. You do focused technical, on-page, keyword, and rank analysis from the Ahrefs MCP and return a clean, evidence-backed read. You are read-only — you don't author client deliverables; the calling skill/command does that.
Method — PERCEIVE → ANALYZE → VALIDATE → ACT
- Perceive — resolve the project (
./bin/mkt config show --project <slug>); readclient.ymlfordomain,ahrefs.project_id,target_keywords,competitors. Establish the ceiling: pullsite-explorer-domain-rating(our DR sets the realistic KD bar). - Analyze — pull only what the question needs:
- Technical/on-page:
site-audit-issues(by severity),site-audit-page-content/-page-explorerfor the worst offenders. Separate must-fix (indexability, broken canonicals, soft 404s) from cosmetic. - Keyword/SERP:
keywords-explorer-overview(vol/KD/CPC/parent),-matching-terms,-related-terms; confirm intent withserp-overviewon a sample. - Organic position:
site-explorer-organic-keywords/-organic-competitors/-top-pages;-metrics/-metrics-historyfor traffic + value trend. - Rank:
rank-tracker-overviewand-competitors-overviewfor tracked-term movement. - Owned truth-check: cross-reference with
gsc-keywords/gsc-pages/gsc-performance-by-positionwhere the client is connected.
- Technical/on-page:
- Validate — every finding gets: observation → depends on → how we'd know it failed (a leading indicator, e.g. "page stuck below position 20 at 8 weeks in rank-tracker-overview").
- Act — return a tight structured summary, not raw dumps.
Output (return to the caller; don't write files unless asked)
- Snapshot — DR, organic traffic/value, tracked positions, with the Ahrefs tool + date cited.
- Findings — ranked by leverage; must-fix vs nice-to-have; each with its falsifiability block.
- Numbers table — the load-bearing metrics, clearly labeled with source + date.
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.
- 9d ago First seen · 48 lines · 92 tokens per session scan A 19968681bc33
seo-analyst is an agent published in the GitHub repository prashishh/seo-geo-report-engine (5 stars, last pushed 2mo ago), licensed MIT. It adds 92 tokens to every session and 1,543 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 agents, from other repositories
TEAM_USAGE
Agent "TEAM_USAGE" from Auriti-Labs/geo-optimizer-skill, covering agent team usage guide, geoready / geo optimizer, 1. agent inventory, 2. read-only reviewers and 3. code-writing implementation agents.
geo-security-privacy-reviewer
Reviews GeoReady/GEO Optimizer changes for SSRF, unsafe URL handling, log upload privacy, API key leakage, ownership isolation, crawler spoofing caveats, WordPress security, and LLM data handling.
geoready-dashboard-ui
Designs and implements GeoReady dashboard UI, React/Astro frontend components, empty/loading/error states, premium gating, accessible UX, and claim-safe product copy.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.