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-cross-account-role-trust-boundariesgit 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-cross-account-role-trust-boundaries)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/auditing-cross-account-role-trust-boundaries"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-cross-account-role-trust-boundaries/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-cross-account-role-trust-boundaries"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/auditing-cross-account-role-trust-boundaries.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 | $0.00173 | $0.01957 |
| Opus 5 | $0.00086 | $0.00979 |
| Sonnet 5 | $0.00035 | $0.00391 |
| Haiku 4.5 | $0.00017 | $0.00196 |
Grade A, and why
auditing-cross-account-role-trust-boundaries 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 — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing cross-account role trust boundaries: when the trust policy lets the wrong caller in
A role's trust policy decides who may assume it, and in a multi-account or partner setup that policy is the whole boundary between organizations. It fails in quiet ways. A principal element written as a wildcard, or as an entire account rather than a specific role, lets any identity in that account assume the role. A third-party integration role that omits or does not verify an external ID is open to the confused-deputy attack, where the vendor is tricked into assuming your role on an attacker's behalf. And assumption chains, where one role assumes another, can carry a low-privilege origin into high-privilege territory the trust policies never meant to connect. You audit these by reading each trust policy for who it actually admits and by following where an admitted principal can go next.
When to use
- Roles in one account can be assumed from another account, a partner organization, or a third-party vendor.
- A trust policy uses a broad principal, or a third-party role relies on an external ID for scoping.
- One assumed role can assume another, forming a chain that may reach privileges the origin should not.
Scope check
Test role assumption only in accounts and organizations you own or are authorized to assess. A confirming assumption obtains real credentials in the target account, so stay inside the authorized accounts and never assume a role belonging to an organization outside your engagement. If you can't name the authorization, stop.
The loop
- Establish the intended trust relationship first. For each cross-account role, name who is supposed to assume it and why: which specific partner role, which service, which internal account. This is the false-positive killer: a trust policy that admits exactly the intended principal under the intended conditions is correct even in a sprawling multi-account setup. Name the intended relationship, then read the policy against it.
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 · 137 lines · 173 tokens per session scan A 53c6392d7c6d
auditing-cross-account-role-trust-boundaries is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 4d ago), licensed MIT. It adds 173 tokens to every session and 1,957 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
cloud-security-engineer
Cloud Security Engineer role bundle for cloud security posture management, IaC review, container security, and identity governance across AWS, Azure, and GCP. Orchestrates cloud-specific posture reviews and zero trust program assessments. Auto-invoked when the user needs help with cloud security configuration, IaC…
azure-review
Performs an Azure security posture review against the CIS Microsoft Azure Foundations Benchmark v2.1.0. Auto-invoked when reviewing Azure infrastructure, Entra ID configurations, NSG rules, Defender for Cloud settings, or Key Vault access policies. Walks through all nine benchmark sections, evaluates each…
gcp-review
Performs a GCP security posture review against the CIS Google Cloud Platform Foundation Benchmark v2.0.0. Auto-invoked when reviewing GCP infrastructure, IAM bindings, VPC firewall rules, Cloud Audit Logs, or GCS bucket security. Walks through all seven benchmark sections, evaluates each recommendation, and produces a…
aws-review
Performs an AWS security posture review against the CIS Amazon Web Services Foundations Benchmark v3.0.0. Auto-invoked when reviewing AWS infrastructure, IAM policies, S3 configurations, CloudTrail settings, VPC security groups, or RDS encryption. Walks through all five benchmark sections, evaluates each…
cloud-pivot-finder
Maps cloud infrastructure from domains and identifies pivot paths from external to cloud internals. Detects cloud providers, enumerates S3/GCS/Azure storage, finds subdomain takeover opportunities, discovers serverless functions, CI/CD exposure, and IaC leaks. Use when user asks for "cloud security", "S3 enumeration"…
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…