qual-doctor
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources to diagnose and tune agent configurations, creating a potential surface for indirect instructions to influence agent tuning.
- Ingestion points: Prospect test cases, CRM data, and raw answers are processed as described in references/task-method.md.
- Boundary markers: There are no specific delimiters or instructions to ignore embedded commands within the ingested prospect data.
- Capability inventory: The skill uses tools to perform workspace mutations and update agent configurations, as noted in SKILL.md and references/task-method.md.
- Sanitization: No explicit validation, filtering, or sanitization of input data is documented before it influences tuning logic.
Audit Metadata