create-feature-flag
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting untrusted user data to define feature flag configurations in a remote system.
- Ingestion points: User-supplied input for flag names, descriptions, traffic types, treatment values, and tags are processed and passed to creation tools in SKILL.md.
- Boundary markers: The instructions include a mandatory 'STOP and confirm' step in Phase 4, requiring explicit human review of the plan before execution.
- Capability inventory: The skill utilizes
harness_create(MCP) andharness create(CLI) to perform write operations within the Harness platform. - Sanitization: The skill relies on backend API validation of identifiers rather than implementing specific sanitization or escaping of the natural language inputs.
Audit Metadata