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 dimitrilaouanis-tech/onyx-mcp --skill verify-before-checkoutgit clone --depth 1 https://github.com/dimitrilaouanis-tech/onyx-mcpWrote 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/dimitrilaouanis-tech/onyx-mcp/verify-before-checkout)<a href="https://agentmods.dev/skills/dimitrilaouanis-tech/onyx-mcp/verify-before-checkout"><img src="https://agentmods.dev/badge/skills/dimitrilaouanis-tech/onyx-mcp/verify-before-checkout/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/dimitrilaouanis-tech/onyx-mcp/verify-before-checkout"><img src="https://agentmods.dev/badge/skills/dimitrilaouanis-tech/onyx-mcp/verify-before-checkout.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 | $0.00089 | $0.00846 |
| Opus 5 | $0.00044 | $0.00423 |
| Sonnet 5 | $0.00018 | $0.00169 |
| Haiku 4.5 | $0.00009 | $0.00085 |
Grade A, and why
verify-before-checkout 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 11d 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify Before Checkout
AI-recommended cloned storefronts are a live attack: data-poisoned fake shops surface inside assistant shopping results, and the page's own claims are the only thing most agents check. This skill adds an independent, cryptographically verifiable fact check before money moves.
When to run this skill
- The agent is about to complete a purchase on a domain the user did not type themselves (recommended, searched, or affiliate-linked).
- The quoted price is significantly below the known market price.
- The storefront resembles a known brand but is not on the brand's primary domain.
- The user asks "is this shop safe / real?"
Procedure
-
Collect the facts — one paid call ($0.25 USDC over x402, Base mainnet):
POST https://onyx-actions.onrender.com/v1/onyx_merchant_fact_check { "domain": "<storefront domain>", "brand": "<brand it appears to represent, if any>", "expected_price": <quoted price, if any> }The response is a signed envelope of raw observations:
domain_age_days,registrar,tls_cert_age_days,tls_issuer,redirected_off_domain,brand_similarity,lookalike_tokens_present,observed_price,price_deviation_pct— each with its method disclosed in_methodology. -
Verify the signature — free call, no payment required:
POST https://onyx-actions.onrender.com/v1/onyx_attestation_verify { "attested": <the full signed response from step 1> }verified: trueproves the observation is genuine Onyx output and untampered. Reject any envelope that fails verification. -
(Optional) Cross-check the price —
onyx_retail_price_check($0.02) against the brand's primary domain, to groundprice_deviation_pct. -
Decide and disclose. Onyx signs facts, not judgments — the interpretation is yours. Sensible agent policy:
domain_age_days < 90andbrand_similarity > 0.5on a non-primary domain → pause and show the user the facts before paying.price_deviation_pct < -40→ surface the deviation explicitly.redirected_off_domain: true→ treat the final domain as the merchant, re-run step 1 against it.- Whatever you decide, show the user the signed facts you acted on. The envelope is your audit trail — it can be re-verified by anyone later.
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.
- 11d ago First seen · 77 lines · 89 tokens per session scan A 7c308e25f84c
verify-before-checkout is a skill published in the GitHub repository dimitrilaouanis-tech/onyx-mcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 89 tokens to every session and 846 once invoked, about $0.0004 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
agent-payment-stats
Cross-protocol agent-economy payment metrics from Barker's index. x402 (Base) volume is on-chain verifiable; ACP / AP2 / MPP / AP4M figures are self-reported claims. Separates real vs nominal volume by filtering wash/noise sellers. Use when users ask about x402 volume, agent payment stats, agent economy metrics, or…
t2000-earn
Earn USDC as a seller on the t2000 marketplace — claim open jobs, deliver, get paid — from Hermes or any terminal agent with a local t2 wallet. Requires t2000-setup. Buyers (hire / post / escrow rules) use t2000-job; Passport Connect users (no local key) use t2000-connect instead.
mppx
TypeScript SDK for the Payment HTTP Authentication Scheme. Handles 402 Payment Required flows with Tempo, Stripe, and other payment methods. Use when integrating payments or mppx into a client or server application.
merchant-context
Resolve sourced merchant facts before relying on them or starting merchant actions.
agentic-discovery
Discover Services and products through InFlow using the Offering Discovery Protocol. Use when an agent needs to search the directory, inspect a Service's supported catalog operations, browse Collections, find Offerings, inspect full product details, or resolve an Action before enrollment or payment.
claude-code
Add this to your project's CLAUDE.md to give Claude Code access to CardZero payments.