kling-official

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines a standard API integration with the official Kling service. It implements several security best practices, including explicit validation for callback URLs and the requirement of user confirmation before substituting providers or uploading local media.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user prompts and external API responses, which could theoretically contain instructions intended to influence the agent's behavior.
  • Ingestion points: User-supplied prompt, image_url, video_url, and callback_url parameters, as well as task results from the Kling API.
  • Boundary markers: Instructs the use of structured placeholders like <<<image_1>>> to map references stably, which helps distinguish data from instructions.
  • Capability inventory: The skill facilitates network requests (POST/GET to the Kling API) and local file reading (converting local paths to base64 for image transmission).
  • Sanitization: Specifically mandates the validation of callback_url strings to ensure they are absolute HTTP or HTTPS URLs prior to use.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 08:42 AM