manage-idp
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data and interpolates it into resource templates via the
harness_createMCP tool. - Ingestion points: Steps 3, 4, 6, and 7 gather multiple free-text fields from the user (e.g.,
<stack>, service patterns, ADR details, and onboarding roles). - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat user-provided data as literal text, creating a surface where malicious input could influence the generated YAML.
- Capability inventory: The skill uses the
harness_createtool to write templates and configurations to the Harness platform. - Sanitization: There is no evidence of sanitization, validation, or escaping logic applied to the user-provided strings before they are interpolated into the YAML/JSON payloads.
Audit Metadata