novu-design-workflow
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines notification workflow templates that ingest untrusted data via
payload.*variables. - Ingestion points: External data is ingested through the
payloadnamespace as described inreferences/step-conditions.mdandreferences/workflow-templates.md(e.g.,payload.webhookUrl,payload.userId). - Boundary markers: The skill uses
{{...}}interpolation markers within workflow templates to delimit external data from instruction strings. - Capability inventory: The skill documents capabilities to perform
HTTP Requeststeps and deliver notifications across multiple channels (Email, SMS, Push, Chat). - Sanitization: The skill promotes the use of
responseBodySchemafor HTTP steps to validate and strictly define addressable properties from external API responses, mitigating schema confusion risks.
Audit Metadata