manage-slos
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data regarding projects, services, and executions from the Harness platform to generate triage reports and runbooks. This data is interpolated into the agent's context without sanitization or boundary markers, creating a surface for indirect instructions to influence agent behavior.
- Ingestion points: Data retrieved via harness_list and harness_get tools for resource types execution, service, project, and environment in SKILL.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when presenting platform data to the agent.
- Capability inventory: The skill utilizes harness_list and harness_get to interact with the Harness MCP server (SKILL.md).
- Sanitization: The instructions do not specify any validation or filtering for the metadata retrieved from the platform.
Audit Metadata