manage-users

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses dedicated MCP tools (harness_list, harness_get, harness_create, harness_delete) to interact with the Harness platform. These tools are scoped to identity management tasks consistent with the skill's description.
  • [DATA_EXPOSURE]: The skill provides the ability to list account-level users and project-level service accounts. This access is consistent with the skill's primary purpose of identity management within the Harness platform.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts parameters like search_term and configuration body content which originate from user instructions.
  • Ingestion points: User-supplied parameters for search_term, body, org_id, and project_id in SKILL.md.
  • Boundary markers: Absent.
  • Capability inventory: Resource creation and deletion capabilities via harness_create and harness_delete tools.
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 06:18 AM
Security Audit — agent-trust-hub — manage-users