aws-solution-architect
Warn
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts (architecture_designer.py, serverless_stack.py, cost_optimizer.py) via subprocesses to automate architectural workflows.\n- [COMMAND_EXECUTION]: The skill performs high-privilege infrastructure management using 'aws cloudformation', 'cdk', and 'terraform' commands, including the 'delete-stack' operation which can lead to data or resource loss.\n- [DATA_EXFILTRATION]: The skill reads local files (requirements.json, current_setup.json, inventory.json) that contain sensitive architectural requirements and resource inventories.\n- [PROMPT_INJECTION]: The skill processes untrusted data from requirements.json to generate infrastructure templates. Ingestion points: requirements.json (SKILL.md). Boundary markers: The skill instructs the agent to treat input as data only, but lacks technical delimiters for the ingested content. Capability inventory: Subprocess execution of Python scripts and cloud deployment tools (AWS CLI, CDK, Terraform). Sanitization: Includes a manual validation step for recommended patterns but lacks automated content sanitization for the input data.
Audit Metadata