The Agent Skills Directory

[SAFE]: The skill implements a security auditing tool for AWS IAM. It queries a local or managed graph database (SubImage) to identify administrator-equivalent roles, cross-account trust chains, and Identity Center (SSO) permissions. All operations use read-only Cypher queries with row limits (LIMIT 200) to prevent performance issues. The skill explicitly instructs the agent to validate schema before execution and warns against fabricating data. No external network requests, data exfiltration, or credential harvesting patterns were detected. The use of Cypher templates is a standard methodology for graph-based security analysis.

investigate-iam