clay-load-scale
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for processing lead data and webhooks from external sources, which creates a potential surface for indirect prompt injection if the ingested data contains instructions meant to influence the agent.
- Ingestion points: The
leadsdata array andwebhookUrlparameter inSKILL.md(Step 2 and Step 4). - Boundary markers: No specific delimiters or "ignore instructions" warnings are included in the provided architectural templates.
- Capability inventory: The skill enables network requests via
fetchand utilizesBash(node:*)for executing the integration logic. - Sanitization: No data validation or sanitization is implemented in these scaling-focused examples; it is expected that the user implementing these patterns will add appropriate validation logic.
Audit Metadata