create-prototype
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from external documentation and context variables to guide its code generation and file-writing actions.
- Ingestion points: The agent reads user stories and PRD artifacts from the
.agents/flow/directory (e.g.,it_{iteration}_PRD.json) and theuser_storycontext variable. - Boundary markers: Instructions do not define specific delimiters or "ignore embedded instructions" warnings for the agent to use when interpreting the content of these external files.
- Capability inventory: The skill allows the agent to create and modify production code and tests in the working directory, as well as append entries to logs in
.agents/flow/. - Sanitization: There is no explicit requirement for the agent to validate or sanitize the content of the user stories or PRD artifacts before execution.
Audit Metadata