analyzing-customers
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the WebSearch tool to ingest and analyze customer feedback from various online platforms, which creates a potential surface for indirect prompt injection where malicious instructions could be embedded in external data.
- Ingestion points: External content fetched via WebSearch from platforms like G2, Capterra, TrustRadius, Amazon, Reddit, and various community forums as described in SKILL.md.
- Boundary markers: There are no explicit instructions or delimiters defined to guide the agent to ignore or isolate potentially malicious instructions within the fetched data.
- Capability inventory: The skill focus is limited to data analysis, persona development, and journey mapping; it does not request capabilities for direct command execution, network exfiltration of local files, or sensitive file system access.
- Sanitization: The instructions do not specify any validation or sanitization routines for the content retrieved from external sources.
Audit Metadata