inngest

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions and code snippets define mechanisms for ingesting and processing external event data, which could potentially carry malicious instructions if the data source is untrusted.
  • Ingestion points: Event payloads accessed via event.data in functions defined in SKILL.md (e.g., user/signed.up, order/placed, and document/uploaded).
  • Boundary markers: The provided code templates do not include explicit delimiter markers or 'ignore' instructions to isolate untrusted data from the execution logic.
  • Capability inventory: The skill patterns demonstrate capabilities including database interaction (db.users), external service communication (Stripe, Resend, OpenAI), and workflow control flows.
  • Sanitization: There are no explicit validation or sanitization steps shown in the snippets to verify the integrity of the event data before it influences subsequent steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 10:54 AM
Security Audit — agent-trust-hub — inngest