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 auditing-cors-and-cross-origin-trustgit 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/auditing-cors-and-cross-origin-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-cors-and-cross-origin-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-cors-and-cross-origin-trust/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/auditing-cors-and-cross-origin-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-cors-and-cross-origin-trust.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 | $0.00188 | $0.02051 |
| Opus 5 | $0.00094 | $0.01026 |
| Sonnet 5 | $0.00038 | $0.00410 |
| Haiku 4.5 | $0.00019 | $0.00205 |
Grade A, and why
auditing-cors-and-cross-origin-trust 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing CORS and cross-origin trust: when an attacker's origin is trusted
The request origin is fully attacker-controlled: any site the victim visits sets it. Cross-origin sharing is safe only because the server decides, per origin, whether to hand back an authenticated response, and a browser enforces that decision. The bugs are all the same shape: the server trusts the origin it was told. It reflects whatever origin arrived and pairs it with credentials, so any site reads the victim's authenticated data; it matches an allowlist by substring, so a lookalike passes; it treats the origin as proof of who is calling. The client-side mirror is a message handler that acts on a posted message without checking where it came from. You find these by reading the code that builds the allow-origin decision, or handles a cross-window message, and asking whether an attacker-set origin is trusted.
When to use
- The code sets cross-origin response headers, statically or by computing them from the request origin.
- An access, authorization, or request-forgery decision is made by looking at the origin or referer.
- Client code receives cross-window messages and acts on their data.
Scope check
Test cross-origin behavior only against applications you own or are authorized to assess, from a test origin and test accounts. A confirmed credentialed cross-origin read exposes real user data, so coordinate. If you can't name the authorization, stop.
The loop
-
Map the decision points. Find where the allow-origin and allow-credentials headers are set and whether the origin value is static or reflected from the request, where any access or request-forgery logic reads the origin or referer, and where client code registers a cross-window message handler. These are the sources and sinks the rest of the loop examines.
-
Check reflected origin with credentials. The dangerous combination is an allow-origin header set to the arbitrary incoming origin together with allow-credentials enabled, on an endpoint that returns cookie-authenticated data. That lets any site read the victim's authenticated response. A wildcard without credentials on genuinely public data is not this bug.
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 · 140 lines · 188 tokens per session scan A b6f65d68497e
auditing-cors-and-cross-origin-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 3d ago), licensed MIT. It adds 188 tokens to every session and 2,051 once invoked, about $0.0009 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
claude-d3js-skill
This skill provides guidance for creating sophisticated, interactive data visualisations using d3.js.
swiftui-expert-skill
Write, review, or improve SwiftUI code following best practices for state management, view composition, performance, macOS-specific APIs, and iOS 26+ Liquid Glass adoption. Use when building new SwiftUI features, refactoring existing views, reviewing code quality, or adopting modern SwiftUI patterns.
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations and post-launch readiness. 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 citations…
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…