Debug "Access Denied" without guessing which of 47 policies is wrong
Traces the full IAM policy evaluation chain when you hit an Access Denied — identity policies, resource policies, SCPs, permission boundaries — and tells you exactly which policy is blocking and how to fix it.
Create a skill called "Permission Sleuth". Help me debug AWS IAM Access Denied errors and optimize IAM policies: Mode 1 — Debug: Given an Access Denied scenario (principal, action, resource), trace the full policy evaluation: 1. List all attached identity policies (inline + managed) 2. Check resource-based policies on the target resource 3. Check SCPs at each level of the AWS Organization 4. Check permission boundaries 5. Check session policies if applicable 6. Identify which policy is causing the deny and suggest the fix Mode 2 — Optimize: Given a role or user: 1. Analyze attached policies for overpermissions 2. If CloudTrail logs are available, compare granted vs. actually-used permissions 3. Generate a least-privilege policy based on actual usage 4. Explain what access would be removed and the risk of each reduction Explain all policies in plain English alongside the JSON.
AWS IAM policy evaluation involves up to 6 layers of policies interacting.
Debugging "Access Denied" means checking all of them. This skill automates
the detective work.
Find the 0.0.0.0/0 rule that was "temporary" 3 years ago
Audits AWS security groups and K8s network policies for overly permissive rules, unused rules, and conflicts. Cross-references VPC Flow Logs to identify rules that are never hit and can be safely removed.
Find the 200 orphaned accounts with admin privileges nobody knew about
Audits user accounts, service accounts, and access keys across your cloud and identity providers. Finds orphaned accounts, unused access keys, over-privileged service accounts, and users who left months ago but still have access.
A calm checklist when fraud hits
Immediate steps, reporting sequences, and long-term hardening for scams and identity theft. Organized by urgency so you know what to do first.
Verify brand offers before you click anything
Screens incoming partnership offers for phishing and scam patterns: suspicious domains, attachments, fake checks, requests to "connect your account," and vague contracts. Outputs a risk rating and safe verification steps.