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 AtlasOmnia/donna-starter --skill marketplace-purchase-vettinggit clone --depth 1 https://github.com/AtlasOmnia/donna-starterWrote 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/atlasomnia/donna-starter/marketplace-purchase-vetting)<a href="https://agentmods.dev/skills/atlasomnia/donna-starter/marketplace-purchase-vetting"><img src="https://agentmods.dev/badge/skills/atlasomnia/donna-starter/marketplace-purchase-vetting/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/atlasomnia/donna-starter/marketplace-purchase-vetting"><img src="https://agentmods.dev/badge/skills/atlasomnia/donna-starter/marketplace-purchase-vetting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 119 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00072 | $0.02547 |
| Opus 5.5 | $0.00029 | $0.01019 |
| Sonnet 5.5 | $0.00014 | $0.00509 |
| Haiku 4.5 | $0.00007 | $0.00255 |
Grade A, and why
marketplace-purchase-vetting 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- marketplace-purchase-vetting — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Marketplace Purchase Vetting
Use this when the user asks whether a local listing is a scam, "too good to be true," worth looking at, or a good deal. Also use this when he asks you to find options — search/discover candidates, then vet the best ones. The goal is not a generic buying guide; it is a practical risk read with clear next steps.
Two modes: Discovery (find me options) and Vetting (check this listing). Run them in sequence when the user hasn't identified a specific listing.
Mode 1: Discovery — Search for candidates in the area
Run this first when the user says "find me a car/suv/truck for $X in our area" without a specific link. The output is a ranked shortlist of the best live options.
Facebook Marketplace search procedure
Read before a broad or tightly constrained vehicle hunt; it documents radius conversion, ignored filters, regional partitioning, feed virtualization, and dealer/down-payment traps.
- Set up the search URL. Facebook Marketplace uses structured URL parameters:
https://www.facebook.com/marketplace/{city}/vehicles?minPrice={min}&maxPrice={max}&sortBy=creation_time_descend&exact=false
- Default to the user's nearest metro area as the city slug; ask when unclear.
- Set price range with ~$500 buffer below max budget (e.g. $3,000--$4,000 for a $3.5k budget) so the floor filters out junk but the ceiling doesn't miss negotiable listings.
sortBy=creation_time_descendshows newest first — gives the freshest picks.
-
Navigate in a logged-in browser session. Facebook Marketplace requires authentication to show results. Use the browser tool (Chrome CDP). Navigate to the constructed URL.
-
Scrape the results page via CDP
Runtime.evaluate— much faster than clicking each listing:
JSON.stringify(Array.from(document.querySelectorAll(
'a[href*="/marketplace/item/"]'
)).map(a => ({
text: a.innerText.trim(),
url: (a.href.match(/marketplace\/item\/(\d+)/) || [])[1]
})).filter(x => x.text))
This returns all visible listings' title, price, location, item ID as one JSON blob.
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 · 182 lines · 72 tokens per session scan A 3aadf3f0c947
marketplace-purchase-vetting is a skill published in the GitHub repository AtlasOmnia/donna-starter (132 stars, last pushed yesterday), licensed MIT. It adds 72 tokens to every session and 2,547 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
marketplace-purchase-vetting
Use this when the user asks whether a local listing is a scam, "too good to be true," worth looking at, or a good deal. Also use this when he asks you to find options — search/discover candidates, then vet the best ones. The goal is not a generic buying guide; it is a practical risk read with clear next steps.
agent-spending-mandate
Write a spending mandate that lets an AI agent buy things for you within strict limits: what it may buy, from whom, up to how much per purchase and in total, until when, and what needs your yes. Use when asked to let my agent make purchases, set a budget for an AI shopping agent, allow agentic checkout, limit what my…
vc-finder
Takes a startup product URL or description, detects the industry and funding stage, identifies 5 comparable funded companies, searches who invested in those companies (Track A), finds VCs who publish investment theses about this space (Track B), and returns a ranked sourced list of relevant investors with deep-dives…
vc-curated-match
Accepts a product description and URL to algorithmically identify relevant Venture Capital investors targeting exactly that stage, industry, and niche based on a curated static dataset.
saas-billing
Implement and audit SaaS billing systems, subscription state machines, secure webhooks, and local database synchronization / Implementasi dan audit sistem billing SaaS, state machine langganan, webhook aman, dan sinkronisasi database lokal.
payment-gateway-expert
Expert guide for integrating payment gateways (Stripe, PayPal, Xendit, Midtrans, DOKU SNAP BI) and secure webhooks into SaaS platforms / Panduan ahli integrasi payment gateway dan webhook aman.