benchmark-methodology
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to browse and analyze third-party external content to perform benchmarking.
- Ingestion points: Instructions in the 'How to collect the data' section of SKILL.md guide the agent to visit competitor websites, review directories such as Clutch.co, LinkedIn profiles, and various portfolio platforms.
- Boundary markers: The instructions lack specific boundary markers or 'ignore embedded instructions' warnings for the agent when processing external site content.
- Capability inventory: The process requires the agent to use web browsing capabilities to access and extract data from untrusted external domains.
- Sanitization: There are no instructions for the agent to sanitize or filter potential malicious prompts embedded in the external content (e.g., hidden text on a competitor's site).
Audit Metadata