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/prashishh/seo-geo-report-enginenpx agentmods add skills/prashishh/seo-geo-report-engine/first-scan-reportWrote 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/prashishh/seo-geo-report-engine/first-scan-report)<a href="https://agentmods.dev/skills/prashishh/seo-geo-report-engine/first-scan-report"><img src="https://agentmods.dev/badge/skills/prashishh/seo-geo-report-engine/first-scan-report/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/prashishh/seo-geo-report-engine/first-scan-report"><img src="https://agentmods.dev/badge/skills/prashishh/seo-geo-report-engine/first-scan-report.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 | $0.00156 | $0.02330 |
| Opus 5 | $0.00078 | $0.01165 |
| Sonnet 5 | $0.00031 | $0.00466 |
| Haiku 4.5 | $0.00016 | $0.00233 |
Grade A, and why
first-scan-report 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 12d 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
first-scan-report
The hook that acquires the client. One scan, one PDF: where they stand, who is beating them, where the
open lanes are, and the plan to take them. It reads the scan artifacts a project already has and
assembles a report.yml the proposal renderer turns into a branded, charted PDF (HTML -> Chrome).
The recipe stays internal; the client sees the finding and the plan, never the method.
The pipeline (run the cheap scans, then assemble)
# 1. multi-source authority (client + competitors)
python3 tools/seo/authority.py --project <slug> --out projects/<slug>/data/raw/dataforseo/authority-<date>.json
# 2. keyword opportunity + content gap
python3 tools/seo/keyword_research.py --project <slug> --location <code> --topic <tokens> \
--out projects/<slug>/data/raw/dataforseo/kw-<date>.json
# 3. multi-engine AI Share of Voice (build a weighted prompt set first, then scan)
python3 tools/geo/prompt_set.py --project <slug> --location <code> --out projects/<slug>/data/prompt-set.json
python3 tools/geo/llm_visibility.py --project <slug> --location <code> --prompt-set projects/<slug>/data/prompt-set.json \
--out projects/<slug>/data/raw/dataforseo/llm-visibility-<date>.json
# 4. assemble + render the hook PDF
python3 tools/report/first_scan.py --project <slug> --date <date> --render
Output: projects/<slug>/deliverables/<slug>-first-scan.pdf (+ the editable first-scan-report.yml).
Every section is optional: a partial scan still renders a coherent report (missing inputs are skipped).
Power sections (the wow: competitive gap + greenfield + AI matrix) — US/Labs markets
# competitor steal-list + traffic-value gap ("you leave $X/mo on the table")
python3 tools/seo/competitor_gap.py --project <slug> --location 2840 --topic <core tokens> \
--out projects/<slug>/data/raw/dataforseo/competitor-gap-<date>.json
# greenfield 2x2: score every candidate keyword Demand x Winnability x Fit, buyer-intent filtered
python3 tools/seo/opportunity_score.py --project <slug> --location 2840 --fit <core tokens> \
--out projects/<slug>/data/opportunity-scored.json
# AI-visibility MATRIX: ~8 buyer questions x 5 engines (named/cited/absent heatmap)
python3 tools/geo/llm_visibility.py --project <slug> --location 2840 --brand "Name" name.com \
--prompts "buyer question 1" "buyer question 2" ... \
--out projects/<slug>/data/raw/dataforseo/ai-matrix-<date>.json
# what competitors win with (their top pages by traffic) - saved to data/raw/.../top-pages-<date>.json
# competitor momentum: funding + hiring direction + news (PredictLeads) - the freshest intel + greenfield proof
python3 tools/seo/competitor_momentum.py --project <slug>
# Google ad activity (Ads Transparency) - who is buying attention
python3 tools/seo/google_ads.py --project <slug>
# social follower gap (needs client.yml social.linkedin URLs + APIFY_API_TOKEN)
python3 tools/seo/social_footprint.py --project <slug>
# traffic share ("you are X% of category traffic") + Core Web Vitals vs rivals
python3 tools/seo/traffic_share.py --project <slug>
python3 tools/seo/web_vitals_bench.py --project <slug> # Core Web Vitals via DataForSEO Lighthouse (no Google key)
python3 tools/seo/reputation.py --project <slug> # reputation + VoC scorecard (ratings + mined complaint themes)
python3 tools/seo/share_of_search.py --project <slug> # share-of-search (branded-demand momentum, Google Trends)
python3 tools/seo/ad_messaging.py --project <slug> # ad-messaging teardown (angles rivals run)
# populate the INTERNAL dashboard (research/competitive-scan.md + BUILD-STATUS.md from all artifacts)
python3 tools/report/scan_summary.py --project <slug> --date <date>
Reports vs dashboard: the PDF is the client deliverable; the dashboard (./bin/mkt dashboard serve)
is the internal live view. Always run scan_summary.py after a scan so the dashboard shows the latest
data (authority, traffic share, AI-visibility, momentum, Core Web Vitals, social, ads) as readable
markdown + CSV. See knowledge/report-roadmap.md for the full section grammar and build order.
When these artifacts exist, first_scan.py auto-adds: The traffic and revenue you are not capturing
(steal list + traffic-value gap), Where rivals are spending and hiring (funding + hiring direction,
the greenfield proof that only N of the field's open roles are in your open lane), Greenfield
opportunity map (bubble matrix), What competitors are winning with (top pages + content pattern),
Where the field is buying attention (Google ad counts), the social presence gap (LinkedIn
followers), and the AI visibility matrix (question x engine heatmap). These turn an audit into a
pitch. Charts: bubble and heatmap in tools/charts/svg.py. Keep the traffic-value number framed as
"competitors combined" (it sums the set) so it stays honest, and reconcile PredictLeads funding totals
against the public record (its coverage can be incomplete). For a fully bespoke, curated deck (like the
Vector Agents pitch), hand-author a _build.py in the project that reads these same artifacts; for a
standard client report, run the pipeline above and let first_scan.py assemble it. See the
/client-report command for the end-to-end runner.
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.
- 12d ago First seen · 130 lines · 156 tokens per session scan A 22c27048d2a8
first-scan-report is a skill published in the GitHub repository prashishh/seo-geo-report-engine (5 stars, last pushed 2mo ago), licensed MIT. It adds 156 tokens to every session and 2,330 once invoked, about $0.0008 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
content-amplifier
Use when the user asks to "amplify influencer content with paid media", "set up whitelisting or Spark Ads", "decide which posts to boost", "repurpose influencer content", "turn one video into multiple ads", or "build a UGC asset library"; produces (paid mode) a content-selection scorecard, a paid amplification…
email-sequence-designer
Use when the user asks to "design a welcome flow", "set up an abandoned-cart sequence", "build a light re-engagement branch inside a lifecycle flow", or "plan a cold-outbound sequence"; produces general lifecycle automation flows (welcome, cart, browse-abandon, post-purchase, in-flow re-engagement, B2B cold outbound)…
reactivation-specialist
Use when the user asks to "build a win-back campaign", "re-engage lapsed subscribers", "run a re-permission / re-consent sweep", or "sunset my dead list"; produces a closed-loop reactivation program — a lapsed-cohort definition, a staged offer ladder, a re-consent (re-permission) capture step, and a sunset-confirm /…
campaign-planner
Use when the user asks to "plan an influencer campaign", "build a campaign blueprint", "track or close a creator campaign", or "record a late campaign correction"; produces the plan and, when requested, a non-canonical evidence tracker with scoped identity, publication, reconciliation, close, and reopen receipts. Not…
product-feed-optimizer
Use when the user asks to "optimize my Shopping feed", "fix product disapprovals", "improve product titles/attributes", or "build feed-driven PMax asset groups"; audits and rewrites the Shopping/Performance Max product feed — title/description patterns, required and recommended attributes, GTIN/availability/price…
cold-outbound-sequencer
Use when the user asks to "build a B2B cold-outbound sequence", "design reply-triage branching", "plan a domain warmup / sending throttle", or "make my outbound CAN-SPAM / opt-in compliant"; produces a multi-step outbound sequence with reply-triage branches (positive / objection / referral / not-now / opt-out), a…