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-group-policy-and-sysvol-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-group-policy-and-sysvol-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-group-policy-and-sysvol-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-group-policy-and-sysvol-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-group-policy-and-sysvol-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-group-policy-and-sysvol-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 | $0.00176 | $0.02153 |
| Opus 5.5 | $0.00070 | $0.00861 |
| Sonnet 5.5 | $0.00035 | $0.00431 |
| Haiku 4.5 | $0.00018 | $0.00215 |
Grade A, and why
auditing-group-policy-and-sysvol-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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing group policy and SYSVOL trust: when a writable policy runs on every machine
Group policy is a domain-wide instruction stream: a policy object names settings, scripts, scheduled tasks, and software, its files live on a share every domain machine reads, and its links decide which machines apply it. Machines apply that policy with high privilege and largely without questioning it. That is safe only when nobody below a domain administrator can change what a policy says, where its files live, or which machines it targets. When a low-privileged principal can write a policy object, edit its files on the share, or alter its link, they run their own scripts and tasks on every machine in scope. A second failure hides in the content itself: a credential stored in a policy file that every domain user can read, or an unsigned policy file a machine applies without verifying its origin. The bug is trust in policy content or scope that a non-administrator can influence. You audit these by reading who can write each policy, its share files, and its links, and what the content discloses.
When to use
- Policy objects, their files on the domain share, or their links may be writable by non-administrators.
- A policy file on the share may carry a stored credential readable by any domain user.
- Policy files may be applied without verifying their origin or may be world-readable on the share.
Scope check
Audit policy and share trust only in directories you own or are authorized to assess, on non-production or a lab domain, making only reversible changes to test policy objects you are permitted to modify and never altering a production policy. A confirmed write runs code on every targeted machine, so keep every probe within scope. If you can't name the authorization, stop.
The loop
- Establish who can actually write the policy, its files, and its link first. For each policy object, read the permissions on the object, on its files on the domain share, and on its links, and determine whether a low-privileged principal holds a write to any of them. This is the false-positive killer: a policy whose object, share files, and links are writable only by administrators cannot be turned against the machines it applies to, however powerful its content. Name the specific write before claiming a path.
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 · 138 lines · 176 tokens per session scan A 750f75d93bf5
auditing-group-policy-and-sysvol-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 176 tokens to every session and 2,153 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 2026-09-09.
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.