startup-pressure-test
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user content, such as startup ideas and target customer profiles, which are then used as the basis for evaluation reports. Because the skill's instructions authorize the agent to use browsing tools for market verification, there is a theoretical vulnerability surface for indirect prompt injection.
- Ingestion points: Startup idea and target customer descriptions described in SKILL.md and playbooks.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt templates.
- Capability inventory: The skill rules in SKILL.md explicitly allow the agent to use browsing tools to verify market facts.
- Sanitization: No input validation, filtering, or sanitization logic is specified for the user-supplied data.
Audit Metadata