build-agent-js
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts on external data by processing requirements (REQ-XXXX) to generate application code, configuration files, and test suites. This creates a surface where malicious instructions embedded in requirement documents could influence agent output.
- Ingestion points: Requirements linked via REQ IDs in the Build Manifest and per-file traceability headers.
- Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined within this skill to isolate processed requirements from the agent's internal logic.
- Capability inventory: The agent performs file system writes (generating artifacts), executes build tools (tsc, vite), and runs testing frameworks (Jest, Vitest, Playwright, Cypress).
- Sanitization: While the skill mandates Zod validation for the code it generates, it lacks explicit sanitization or validation protocols for the requirements it ingests as input.
Audit Metadata