compliance-audit
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides extensive Bash command templates in 'references/evidence-collection.md' for the agent to execute. These include commands to export IAM details, security group rules, and encryption settings from AWS, GCP, and Azure environments, as well as running security scanners like Checkov and Semgrep.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it is designed to ingest and analyze external system data that could be influenced by an attacker.\n
- Ingestion points: The skill reads cloud configuration exports, application access control lists, and security scan results as described in 'references/evidence-collection.md'.\n
- Boundary markers: There are no instructions or delimiters provided to the agent to distinguish between the content of the evidence being collected and the skill's operational instructions.\n
- Capability inventory: The skill utilizes the 'Bash' tool for command execution and the 'Write' tool for artifact storage, which could be abused if an injection occurs.\n
- Sanitization: The skill documentation does not specify any sanitization, filtering, or validation steps for the data collected from external APIs and logs before it is processed.
Audit Metadata