agent-generator
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill generates agent definition files based on user-provided parameters, creating an attack surface where malicious input could be persisted into project artifacts.\n
- Ingestion points: User inputs for agent 'name', 'description', and 'expertise' fields processed in
SKILL.md.\n - Boundary markers: No delimiters or instructions to ignore embedded commands are present in the instructions to prevent the model from obeying injected content.\n
- Capability inventory: The skill is configured with 'Write' and 'Edit' tools in
SKILL.mdto modify the local filesystem.\n - Sanitization: No input validation, escaping, or filtering logic is specified to handle potentially malicious input strings before they are written to files.
Audit Metadata