kling-official
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided prompts and media references which are then sent to the Kling API. This creates a surface for indirect prompt injection. Ingestion points:
prompt,image_url, andvideo_urlparameters inSKILL.md. Boundary markers: Absent. Capability inventory: Performs API calls to Kling and reads local files for conversion. Sanitization: None described for user prompt inputs. - [DATA_EXFILTRATION]: The skill reads local media files (e.g., via
image_pathorsound_file) and transmits them to the official Kling API endpoint atapi-singapore.klingai.com. This is the intended function of the skill but involves sending local data to an external service provider.
Audit Metadata