workflows-claygent
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from external lead searches or manual inputs into agent prompts via template variables.
- Ingestion points: Records from "Find leads", "Audiences", or manual triggers are used as inputs for downstream agent nodes in SKILL.md.
- Boundary markers: The instructions lack explicit requirements for using delimiters or adding safety constraints when interpolating record data into the "{{variable}}" syntax within agent prompts.
- Capability inventory: The created agent nodes are intended to perform "per-record web research" and output structured data.
- Sanitization: No specific sanitization or validation of the input record content before prompt interpolation is described in the workflow setup.
Audit Metadata