business-knowledge-workflow

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Indirect Prompt Injection (SAFE): The skill is designed to ingest data from internal iWiki pages and local source code files.
  • Ingestion points: Tools such as iWiki.getDocument, search_content, and read_file are used in SKILL.md to pull external data into the agent context.
  • Boundary markers: Not explicitly defined in the workflow instructions.
  • Capability inventory: Includes search_content and read_file for internal code analysis.
  • Sanitization: None specified, as the skill relies on the integrity of internal corporate resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:49 PM