agent-designer
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of documentation and local Python scripts for system design and evaluation. Analysis of the scripts (agent_planner.py, tool_schema_generator.py, agent_evaluator.py) confirms they only perform data processing on provided JSON files and do not engage in network activity, shell command execution, or sensitive data access.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external JSON data. 1. Ingestion points: assets/sample_system_requirements.json, assets/sample_tool_descriptions.json, and assets/sample_execution_logs.json. 2. Boundary markers: Absent in script outputs. 3. Capability inventory: No subprocess, exec, or network operations. 4. Sanitization: Absent. While this presents a surface for indirect prompt injection, the scripts themselves do not possess exploitable capabilities such as network access or shell execution.
Audit Metadata