implement
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external 'Ticket' and 'Specification' data which serves as a potential surface for indirect prompt injection.
- Ingestion points: Step 1 and 2 in
SKILL.mdrequire the agent to read and bind to canonical Ticket and Specification revisions from the external Workflow context. - Boundary markers: The instructions focus on binding to exact revisions, scope, and acceptance criteria, but lack explicit guardrails or instructions for the agent to ignore or isolate commands embedded within the ticket content itself.
- Capability inventory: The skill provides the agent with capabilities to perform direct execution, delegate bounded actions to child 'Executors', and invoke 'Supplemental skills' at specific extension points.
- Sanitization: There is no mention of sanitization, filtering, or validation of the content within the tickets or specifications before they are used to guide execution.
Audit Metadata