ai-on-saas-discovery-and-qualification
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill requires the agent to analyze untrusted external data such as tender documents and architecture records to generate proposal content, creating a surface for indirect prompt injection.
- Ingestion points: Tender documents, discovery evidence, and architecture records processed during the qualification workflow.
- Boundary markers: The instructions lack explicit delimiters or directions to ignore embedded commands within ingested data.
- Capability inventory: The skill is authorized to read workspace files and write to BRIEF.md and other proposal artifacts.
- Sanitization: No instructions for sanitizing or validating input data are present.
Audit Metadata