eng-lead
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input to determine which engineering tools and skills to invoke, such as code generation or system design. This interaction model presents a surface for indirect prompt injection where malicious instructions could be embedded in user requests.
- Ingestion points: Processes all user messages to determine intent (Phase 1 in SKILL.md).
- Boundary markers: Implements a mandatory clarification phase using the AskUserQuestion tool to bound the request scope before any tool execution.
- Capability inventory: Orchestrates a wide array of skills including product-spec, system-design, go-feature, and db-migrate (Phase 2 Routing Table).
- Sanitization: Utilizes structured selection options within the AskUserQuestion tool to interpret user intent instead of direct interpolation of raw user text into tool arguments.
- [NO_CODE]: The skill consists entirely of instructional markdown and orchestration logic without embedded scripts or executables.
Audit Metadata