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 reviewing-content-security-policygit 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/reviewing-content-security-policy)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/reviewing-content-security-policy"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/reviewing-content-security-policy/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/reviewing-content-security-policy"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/reviewing-content-security-policy.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 | $0.00177 | $0.02020 |
| Opus 5.5 | $0.00071 | $0.00808 |
| Sonnet 5.5 | $0.00035 | $0.00404 |
| Haiku 4.5 | $0.00018 | $0.00202 |
Grade A, and why
reviewing-content-security-policy 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 — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reviewing a content security policy: whether it stops the injection or only looks strict
A content security policy is a defense-in-depth control against injected script, and reviewing one is less about listing weak directives than about a single judgment: if an attacker gets markup onto this page, does the policy stop the script from running. That reframing matters both ways. A permissive policy on a page with a real injection vector turns a mitigated cross-site-scripting bug into an exploitable one, and is a genuine finding. The same permissive policy on a page with no injection sink is defense-in-depth worth hardening but not an exploit. And a policy that looks weak, an inline allowance sitting next to a nonce, is often not weak at all, because the browser ignores the legacy allowance when the nonce is present. You review it by reading the policy against the pages it protects and asking what an injected script could still do.
When to use
- A response sets a content security policy, in a header, a meta tag, or configuration.
- You are already looking at a page that reflects or stores user input, or want to know whether one is protected.
- You need to tell a policy that blocks injected script from one that only appears to.
Scope check
Review policies on applications you own or are authorized to assess, and demonstrate a bypass only against pages in scope. A confirmed policy bypass paired with an injection is a working cross-site script, so treat it as such and coordinate. If you can't name the authorization, stop.
The loop
-
Map the policy, its delivery, and the sinks it must protect. Read the policy and how it is delivered, and find whether the pages it covers have an injection sink: a place user input reaches markup or script. The presence or absence of that sink decides whether a weakness is exploitable or only defense-in-depth, so establish it first.
-
Check inline script. A script source that allows inline script with no nonce or hash lets an injected inline script or event-handler attribute run. This is the single most common real weakness, but read it carefully: a nonce or hash present alongside the inline allowance makes a modern browser ignore the allowance, so that combination is mitigated, not broken.
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 · 134 lines · 177 tokens per session scan A 26db5aa80fff
reviewing-content-security-policy is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 177 tokens to every session and 2,020 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.
varlock-claude-skill
Secure environment variable management ensuring secrets are never exposed in Claude sessions, terminals, logs, or git commits.