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-third-party-script-and-sri-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-third-party-script-and-sri-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-third-party-script-and-sri-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-third-party-script-and-sri-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-third-party-script-and-sri-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-third-party-script-and-sri-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 | $0.00207 | $0.02332 |
| Opus 5.5 | $0.00083 | $0.00933 |
| Sonnet 5.5 | $0.00041 | $0.00466 |
| Haiku 4.5 | $0.00021 | $0.00233 |
Grade A, and why
auditing-third-party-script-and-sri-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 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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing third-party script and SRI trust: a script tag is code you did not write running as you
Every external script a page loads runs in that page's security context: it can read the DOM, cookies accessible to script, form fields, and tokens, and it can send them anywhere. So a third-party script is not a dependency the way a backend library is; it is code you did not write, delivered by a party you do not control, executing with your page's full privileges on your users. The risk is supply-chain: the vendor's CDN is compromised, the vendor ships a malicious update, or the script URL is swapped, and altered code runs on every visitor. A form-skimming attack that steals payment details is exactly this, a legitimate analytics or widget script replaced with one that also reads the checkout form. The defenses are integrity pinning (a Subresource Integrity hash so the browser refuses altered code) and a Content-Security-Policy that constrains which origins may serve script at all, and a tag manager that injects further scripts undoes both by loading code nobody reviewed. The audit inventories every third-party script, checks whether its integrity is pinned and its origin constrained, and treats any unpinned externally served script as attacker-controllable. You audit this by listing what the page loads and asking, for each, what happens if that code is swapped.
When to use
- A web page loads JavaScript from another origin: a CDN, an analytics or marketing tag, a payment or chat widget, a tag manager.
- External script tags may lack Subresource Integrity hashes, or scripts may be served over mutable or wildcard URLs.
- A tag manager or loader may inject further scripts, and the Content-Security-Policy may be missing or permissive.
Scope check
Audit third-party script trust only on sites and properties you own or are authorized to assess. Reviewing what a page loads is passive, but testing swap or injection scenarios touches real third-party relationships, so stay within your own properties and test deployments and never tamper with a vendor's delivery or another site's scripts. If you can't name the authorization, stop.
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 · 147 lines · 207 tokens per session scan A 2038a52a15de
auditing-third-party-script-and-sri-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 207 tokens to every session and 2,332 once invoked, about $0.0008 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.