client-onboarding

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a structured business process for researching companies and generating growth strategies. All operations, including file creation and the use of external research tools, are consistent with the skill's stated purpose.
  • [EXTERNAL_DOWNLOADS]: The playbook utilizes web search, web fetching, and specialized scrapers (Meta, Google, LinkedIn) to gather public intelligence on companies, competitors, and founders. These activities are essential for the research phases described in the skill.
  • [PROMPT_INJECTION]: The skill processes untrusted data from the web during its intelligence-gathering phase. This data is subsequently synthesized into a 'Client Intelligence Package.' This workflow presents an inherent surface for indirect prompt injection, where malicious content hosted on a researched website could attempt to influence the agent's synthesis or strategy generation. However, the skill employs structured frameworks and diagnostic steps that mitigate the likelihood of accidental obedience to such content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 09:59 PM