cloudflare-workers-runtime-apis

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides legitimate documentation and code templates for Cloudflare Workers development. All code samples follow platform standards and no malicious patterns were identified.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides templates for ingesting data from external APIs and WebSocket messages, which constitutes an attack surface. However, the provided examples include robust safety measures:
  • Ingestion points: Data enters the worker context through fetch() responses and webSocketMessage() events in the Durable Object templates.
  • Boundary markers: The templates include checks for response status codes and proper error handling for JSON parsing.
  • Capability inventory: The skill uses network requests, cryptographic operations, and edge storage.
  • Sanitization: Code examples demonstrate manual input sanitization, such as trimming whitespace and limiting the length of user-provided content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:01 AM
Security Audit — agent-trust-hub — cloudflare-workers-runtime-apis