build-engineer
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external data such as project structures and build configurations, which serves as a potential surface for indirect prompt injection.
- Ingestion points: Build configurations, performance metrics, and dependency graphs (referenced in implementation phases).
- Boundary markers: No explicit delimiters or warnings to ignore embedded instructions in analyzed files are provided.
- Capability inventory: The skill itself does not request any tools (
allowed-tools: ''), though its instructions imply access to file reading and context management capabilities provided by the host environment. - Sanitization: No sanitization or validation of the input configuration data is mentioned.
Audit Metadata