inngest-middleware
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing official middleware packages (@inngest/middleware-encryption, @inngest/middleware-sentry) and the Sentry Node SDK. These are verified packages from well-known technology companies and service providers.
- [PROMPT_INJECTION]: The code examples demonstrate patterns for processing untrusted event data and passing it to an LLM, which is a standard surface for indirect prompt injection.
- Ingestion points: Untrusted data enters via the event object in function handlers and middleware (e.g., event.data.content in SKILL.md).
- Boundary markers: No delimiters or instructions are shown to guide the LLM to ignore potentially malicious content within the event data.
- Capability inventory: The skill examples utilize the openai client for network-based LLM requests and PrismaClient for database operations.
- Sanitization: The provided examples do not include explicit validation or sanitization of the input data before it is sent to external services.
Audit Metadata