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 UnboundCompute/security-agent-skills --skill hunting-price-and-coupon-manipulationgit clone --depth 1 https://github.com/UnboundCompute/security-agent-skillsWrote 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/unboundcompute/security-agent-skills/hunting-price-and-coupon-manipulation)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-price-and-coupon-manipulation"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-price-and-coupon-manipulation/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/unboundcompute/security-agent-skills/hunting-price-and-coupon-manipulation"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-price-and-coupon-manipulation.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 | $0.00181 | $0.02180 |
| Opus 5.5 | $0.00072 | $0.00872 |
| Sonnet 5.5 | $0.00036 | $0.00436 |
| Haiku 4.5 | $0.00018 | $0.00218 |
Grade A, and why
hunting-price-and-coupon-manipulation 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.
How it starts
The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting price and coupon manipulation: whoever computes the total decides what you pay
The amount a customer is charged is computed from item prices, quantities, and discounts, and the only safe place to do that computation is the server, from data the buyer cannot change. When any input to the total is taken from the client, the buyer sets the price. The failures are direct. A price, quantity, or line total sent in the request and trusted, rather than recomputed from a server-side catalog, lets a buyer name their own price. A negative quantity subtracts from the total, and a quantity large enough to overflow the arithmetic wraps it to a small or negative number. A discount or coupon that stacks with itself or others, is reused past its intended limit, or applies to items it was never meant for, cuts the total below its floor. And a total computed in the browser and accepted at checkout hands the whole calculation to the attacker. The hunt is to find any input to the charged total that the buyer controls. You hunt this by altering prices, quantities, and coupons in the request and seeing what the server charges.
When to use
- A buyer influences cart contents, quantities, or promotions and the server computes a charged total.
- Prices, quantities, or line totals may be taken from the client request rather than a trusted catalog.
- Coupons or discounts may stack, be reused past their limit, or apply to items outside their scope.
Scope check
Test pricing and promotions only against applications you own or are authorized to assess, on non-production or sandbox checkout paths. Manipulating totals and placing orders exercises real commerce, so use test accounts and sandbox payment and never complete a real underpriced purchase or consume real promotions. If you can't name the authorization, stop.
The loop
- Establish the server-side price of truth first. Confirm the total should be recomputed on the server from a trusted catalog of prices, with server-validated quantities and server-enforced coupon rules, and that no price, line total, or discount from the client is trusted. This is the false-positive killer: a checkout that recomputes every amount server-side from the catalog, bounds quantities, and enforces coupon scope, stacking, and reuse limits is correct. Name the price of truth, then test each input.
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 · 137 lines · 181 tokens per session scan A 9041ce596d35
hunting-price-and-coupon-manipulation is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 181 tokens to every session and 2,180 once invoked, about $0.0007 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 2026-09-05.
Other skills, from other repositories
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations, post-launch readiness and people. Use when asked to check whether a project is ready to ship, to audit an area before launch, or to work through a specific checklist. Produces evidence with file:line…
integrate-arcjet-guard-genkit
Integrate Arcjet security into a Genkit JS agent using @arcjet/guard — wrap ai.defineTool, put guardMiddleware on generate({ use }) for unwrapped / MCP / filesystem tools, and read a caller-owned id from generate({ context }). Use when asked to add Arcjet to genkit, rate limit its tools, screen inbound messages, or…
integrate-arcjet-guard-tanstack-ai
Integrate Arcjet security into a TanStack AI chat() app using @arcjet/guard — put guardMiddleware first on chat({ middleware }) so onBeforeToolCall gates tools, and read a caller-owned id from chat({ context }). Use when asked to add Arcjet to TanStack AI, rate limit its tools, screen inbound messages, or block prompt…
integrate-arcjet-guard-claude-managed-agents
Integrate Arcjet security into Claude Managed Agents (hosted REST+SSE, beta managed-agents-2026-04-01) using @arcjet/guard — screen user.message / initialevents before sessions.events.send, and gate custom tools on agent.customtooluse. Use when asked to add Arcjet to Claude Managed Agents, rate limit custom tools, or…
linear-claude-skill
Manage Linear issues, projects, and teams.
panguard
AI agent security platform — audit skills, scan for threats, and run 24/7 protection with 9,700+ detection rules.