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-break-glass-account-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-break-glass-account-trust)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-break-glass-account-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-break-glass-account-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-break-glass-account-trust"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-break-glass-account-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 | $0.00213 | $0.02528 |
| Opus 5.5 | $0.00085 | $0.01011 |
| Sonnet 5.5 | $0.00043 | $0.00506 |
| Haiku 4.5 | $0.00021 | $0.00253 |
Grade A, and why
auditing-break-glass-account-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 — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing break-glass account trust: the last-resort key must be sealed, watched, and rare
A break-glass account exists for the emergency where normal admin access is broken, the identity provider is down, everyone is locked out, so it holds the highest privilege and, by design, bypasses some of the controls that gate everyone else. That combination, maximum power plus fewer gates, is precisely why it is dangerous when its own controls are loose. The account is supposed to be a sealed last resort: strong unique credential, rotated after each use, covered by logging and alerting, used rarely and reviewed every time. The failures are where it stops being sealed. A static, shared, or never-rotated credential turns the emergency key into a permanent superuser login anyone who once saw it can use. An emergency account excluded from the multi-factor, conditional-access, and logging controls that cover other admins is both easy to use and invisible when used. A just-in-time or privileged-access-management elevation that grants more than the task needs, or is not revoked when the task ends, leaves standing privilege behind a temporary label. An emergency path with no approval, no time bound, and no alerting gets used routinely instead of exceptionally. And a break-glass use that triggers no review means the one time it mattered, nobody looked. The audit checks that the emergency identity is sealed, watched, minimally scoped, and reviewed, so its power is available in a real emergency and inert otherwise. You audit this by asking who could use it right now, whether anyone would notice, and whether it would ever be un-elevated.
When to use
- A system has a break-glass or emergency-access account, or a just-in-time privileged-elevation path, that holds power beyond ordinary admins.
- The emergency credential may be static, shared, or never rotated, or the account may be excluded from MFA, conditional-access, or logging controls.
- An elevation may over-grant, may not be revoked when the task ends, or a break-glass use may trigger no alerting or review.
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 · 156 lines · 213 tokens per session scan A e51f5c6c9968
auditing-break-glass-account-trust is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 213 tokens to every session and 2,528 once invoked, about $0.0009 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.