exploratory-testing
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to interact with and process data from external, untrusted software systems under test.
- Ingestion points: Untrusted data enters the agent context via browser automation capabilities, including page navigation, screenshot capture, and API call monitoring in the SKILL.md file.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for data retrieved from the tested system.
- Capability inventory: The agent has file-write capabilities in the local project directory (e.g., creating 探索笔记_{主题}.md) and uses browser interaction tools as listed in the SKILL.md workflow.
- Sanitization: The skill lacks explicit sanitization or validation steps for the content scraped or intercepted from the system under test.
Audit Metadata