workflow-feature-shipper
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted requirements from a user-supplied file (
feature.md) and uses them to drive code implementation, artifact generation, and tool execution, creating a surface for indirect prompt injection. - Ingestion points: The skill reads and processes the content of
feature.mdin Step 1. - Boundary markers: The instructions do not define delimiters or provide guidance to the agent to ignore potentially malicious instructions embedded within the requirements file.
- Capability inventory: The skill has the capability to modify the repository filesystem, execute verification tools, generate artifacts in the
evidence/directory, and delegate tasks to sub-agents. - Sanitization: No sanitization, validation, or filtering of the requirement file content is performed before the agent acts upon it.
Audit Metadata