fullstack-dev
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes instructions in the mandatory workflow of
SKILL.mdthat guide the agent to execute shell commands such asnpm run buildandcurlfor local health checks. These commands are standard development practices for validating the generated code and verifying that services are running as expected.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for building applications that process external data, creating an ingestion surface.\n - Ingestion points: The agent ingests user requirements and external API specifications to generate scaffolds and business logic.\n
- Boundary markers: The instructions utilize a structured "MANDATORY WORKFLOW" and checklists in
SKILL.mdto guide agent behavior.\n - Capability inventory: The skill employs shell command execution and file operations to implement full-stack services.\n
- Sanitization: The skill explicitly teaches and requires input validation at the application boundary using schemas (e.g., Zod or Pydantic), which mitigates the risk of external data compromising the generated application.
Audit Metadata