aws-diagram
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill generates Python code at runtime and executes it via the shell to produce diagrams. \n
- The process explicitly instructs the agent to "Write the code to a temporary file" and "Execute it with python .py".\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from an external JSON file to generate executable code, creating a surface for potential injection attacks. \n
- Ingestion points: Reads infrastructure data from
aws_infrastructure.jsonin the current directory.\n - Boundary markers: No instructions for using delimiters or guards to separate user data from script instructions.\n
- Capability inventory: File system writes and shell command execution (
python).\n - Sanitization: Instructions do not mandate escaping or validation of the JSON content before inclusion in the Python script.\n- [COMMAND_EXECUTION]: The skill uses the command line to run the generated Python scripts.
Audit Metadata