meta-sales-lead-researcher

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and orchestration logic do not contain malicious patterns, unauthorized network requests, or credential theft.- [PROMPT_INJECTION]: An indirect prompt injection surface is present due to the processing of external search data.
  • Ingestion points: The source_review step in SKILL.md ingests output from the web_search step.
  • Boundary markers: The skill does not utilize explicit delimiters or 'ignore' instructions for data gathered from search engines.
  • Capability inventory: The skill utilizes search tools and sub-agent orchestration. No capabilities for arbitrary code execution or direct file system manipulation were observed in the skill body.
  • Sanitization: While user input is escaped via xml_escape, external search data is interpolated directly without specific sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 12:04 PM
Security Audit — agent-trust-hub — meta-sales-lead-researcher