backend-patterns
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The skill instructions explicitly handle data from untrusted sources including webhooks, import flows, and paste handlers (SKILL.md).
- Boundary markers: There are no specific instructions for the agent to use delimiters or ignore embedded instructions when processing these external data sources.
- Capability inventory: The agent is authorized to perform database writes and execute shell commands for database management (e.g.,
pnpm db:migrate,pnpm db:push). - Sanitization: No specific sanitization or validation logic is defined for the ingestion of external links beyond routing them through a helper function.
Audit Metadata