customer-research
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and analyze untrusted external data.
- Ingestion points: The skill instructions specifically direct the agent to process external materials such as customer interview transcripts, support tickets, and public content from platforms including Reddit, G2, Hacker News, and YouTube.
- Boundary markers: The instructions lack explicit boundary markers or directives to the agent to ignore or isolate potential instructions embedded within the analyzed research materials.
- Capability inventory: The skill provides a framework for the agent to analyze, cluster, and synthesize information into personas and reports. In its execution environment, the agent typically possesses capabilities like web browsing or file access to retrieve and process these materials.
- Sanitization: No sanitization, validation, or filtering mechanisms are described to prevent instructions embedded in external research data from influencing the agent's logic during the synthesis phase.
Audit Metadata