santa-method
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill implements a workflow that interpolates potentially untrusted generator output into the prompts of reviewer agents, which is a known surface for indirect prompt injection.\n
- Ingestion points:
SKILL.md(TheREVIEWER_PROMPTtemplate in Phase 2 interpolates the{output}variable into a reviewer agent prompt).\n - Boundary markers: The prompt template uses Markdown headers (e.g.,
## Output Under Review) to separate variables, which provide limited structural separation against adversarial content.\n - Capability inventory: The skill utilizes the
Agenttool to spawn parallel sub-agents for verification and fixing tasks.\n - Sanitization: No explicit escaping, sanitization, or validation of the generator's output is mentioned in the implementation patterns.
Audit Metadata