icp-persona-builder

Pass

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external websites.
  • Ingestion points: The skill uses WebFetch to read company websites and WebSearch to gather reviews, case studies, and social signals.
  • Boundary markers: There are no explicit instructions to use delimiters or ignore embedded commands when processing fetched web content.
  • Capability inventory: The skill instructs the agent to write research findings and generated personas to the local filesystem in the clients/ directory.
  • Sanitization: No sanitization or filtering of the external text is performed before it is used to influence the agent's output and file-writing actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 28, 2026, 11:41 AM