creating-aws-architecture-diagrams
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it is designed to read and process untrusted data from the local repository.
- Ingestion points: As per the workflow in SKILL.md, the agent reads 'relevant code, IaC, configuration, and architecture notes' to ground the topology.
- Boundary markers: No specific delimiters or safety instructions are used to separate ingested code from the agent's internal instructions.
- Capability inventory: The skill can execute a local Python script for image processing and use an external image generation tool.
- Sanitization: No validation or filtering is performed on the ingested code to prevent the execution of embedded instructions.
- [COMMAND_EXECUTION]: The skill executes a local shell command to resize and verify image artifacts.
- Evidence: SKILL.md provides a command line template to run 'python scripts/ensure_png_size.py' with variable input/output paths and dimensions.
- Mitigation: The included script uses standard library argument parsing and image processing, presenting no evidence of malicious intent or command injection vulnerabilities.
Audit Metadata