explain-flag
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes feature flag metadata and targeting rules from the Harness API. This external data is a potential vector for indirect prompt injection.
- Ingestion points: Flag metadata and definitions fetched via
harness_listandharness_getin Steps 3 and 4 of SKILL.md. - Boundary markers: No specific boundary markers or instructions to ignore embedded content are defined for the retrieved metadata.
- Capability inventory: The skill is limited to read-only tool calls (
harness_get,harness_list) and does not possess capabilities for shell execution, file system modification, or arbitrary network access. - Sanitization: No sanitization of the retrieved text strings is described before they are formatted and presented to the user.
Audit Metadata