site-explorer

Pass

Audited by Gen Agent Trust Hub on Mar 27, 2026

Risk Level: SAFE
Full Analysis
  • Intended Data Interaction: The skill is designed to interact with SharePoint data including site metadata, list items, and file contents via the workiq-ask_work_iq tool. This access is required for the skill's navigation and discovery functionality.\n- External Data Ingestion Surface: The skill processes information retrieved from external SharePoint sites. While this presents a potential surface for indirect prompt injection where data retrieved from SharePoint could contain instructions, this is a common trait of data-exploration tools and is handled by the agent's underlying safety mechanisms.\n
  • Ingestion points: Site names, list schemas, list items, and file contents retrieved via the workiq-ask_work_iq tool.\n
  • Boundary markers: The instructions do not define explicit delimiters or 'ignore' warnings for the data being processed.\n
  • Capability inventory: The agent can execute further workiq-ask_work_iq tool calls and present data to the user based on the content found.\n
  • Sanitization: No explicit validation or filtering of SharePoint content is performed before processing.\n- Use of Local CLI Tools: The skill relies on the workiq CLI to perform queries. This is an expected pattern for integration with local productivity environments and does not involve unauthorized network operations or privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 27, 2026, 10:07 AM