one-get-task-details
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes full task details from an external source (One Horizon MCP). This content is considered untrusted as it could contain malicious instructions designed to subvert the agent's logic.
- Ingestion points: Data returned by the
get-task-detailsMCP tool. - Boundary markers: The instructions do not define delimiters or warnings to ignore instructions within the fetched task content.
- Capability inventory: The skill invokes the
get-task-detailstool to fetch data from the workspace. - Sanitization: There is no evidence of sanitization or validation performed on the task details before they are presented to the agent context.
Audit Metadata