manage-feature-flags

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied strings such as feature flag names, descriptions, and tags which are then interpolated into tool parameters for harness_create and harness_update. While this presents a surface for indirect prompt injection if a user provides instructions within these fields, it is a standard functional requirement for this type of skill.
  • Ingestion points: User-provided flag metadata in Step 3 and Step 6 (SKILL.md).
  • Boundary markers: Absent.
  • Capability inventory: Resource creation, modification, and deletion via harness_create, harness_update, harness_execute, and harness_delete tools.
  • Sanitization: Not explicitly defined within the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 11:24 PM
Security Audit — agent-trust-hub — manage-feature-flags