skills/mukul975/anthropic-cybersecurity-skills/scanning-iac-and-images-with-trivy/Gen Agent Trust Hub
scanning-iac-and-images-with-trivy
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The helper script
scripts/agent.pyexecutes thetrivybinary usingsubprocess.runto perform security scans. - Evidence: Found in
scripts/agent.pywhere therun_scanfunction executes a command list. The implementation mitigates shell injection risks by passing arguments as a list and not usingshell=True.- [PROMPT_INJECTION]: The skill facilitates the ingestion of data from external, potentially untrusted sources (container images, remote Git repositories), which introduces a surface for indirect prompt injection. - Ingestion points: The scan results from
trivy(processed inscripts/agent.py) include data derived from external files, Docker layers, and repository contents. - Boundary markers: Absent. Scan findings such as vulnerability IDs, package names, and misconfiguration titles are returned to the agent context without delimiters or safety instructions.
- Capability inventory: The skill can execute the
trivybinary via subprocess and write output files to the local filesystem. - Sanitization: There is no sanitization or escaping of the metadata retrieved from the scanned targets before it is presented to the agent.
Audit Metadata