aws-security

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the AWS CLI to perform security checks, including aws sts get-caller-identity, aws iam list-account-aliases, and aws ec2 describe-security-groups.
  • [DATA_EXFILTRATION]: The skill accesses highly sensitive data by generating and decoding the IAM credential report (aws iam get-credential-report). While it displays the content locally for auditing, this involves handling raw credential metadata. This behavior is consistent with the skill's primary purpose.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes resource names and metadata from an AWS environment that could be controlled by an external actor.
  • Ingestion points: AWS API responses for IAM users, roles, and EC2 resource descriptions in SKILL.md.
  • Boundary markers: Absent; resource data is processed without delimiters or instructions to ignore embedded commands.
  • Capability inventory: Extensive read access to IAM, EC2, and Cloud Control APIs, along with Python execution capabilities.
  • Sanitization: Absent; resource attributes are directly interpolated into output or processed in loops.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 03:29 AM
Security Audit — agent-trust-hub — aws-security