zhipu-embedding

Warn

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The Python code example references a library named zai (from zai import ZhipuAI). The official SDK for Zhipu AI is zhipuai. Referencing an alternative or potentially misspelled package name can lead to the installation of unverifiable dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data to generate embeddings.
  • Ingestion points: The input parameter in the client.embeddings.create call accepts user-provided text strings (documented in SKILL.md).
  • Boundary markers: The provided instructions do not include specific delimiters or warnings to ignore instructions embedded within the text intended for embedding.
  • Capability inventory: The skill utilizes network operations to send data to https://open.bigmodel.cn/api/paas/v4/embeddings (documented in SKILL.md).
  • Sanitization: There are no documented steps for filtering or sanitizing the input text before it is processed by the model.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 3, 2026, 07:15 PM
Security Audit — agent-trust-hub — zhipu-embedding