nestjs-professional-software-engineering
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted codebase content to perform its engineering tasks.
- Ingestion points: The workflow (SKILL.md) instructs the agent to read project implementation, callers, tests, configuration, manifests, and repository instructions.
- Boundary markers: There are no specific instructions to ignore malicious natural language prompts embedded within the files it reads.
- Capability inventory: The skill is authorized to edit files and execute project-defined commands such as tests, builds, and linting.
- Sanitization: While the skill requires validating untrusted input for the software it writes, it does not specify sanitization for the content it reads during inspection.
Audit Metadata