byted-web-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill prompt instructs the agent to ask the user to paste their API key directly into the chat box ("API Key 直接在本聊天框发给我即可"), which is a pattern where the LLM handles and receives raw sensitive secret values.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The required workflow uses a web search tool (byted-web-search) to fetch arbitrary web pages based on user queries, which exposes the agent to outsider-authored free text via search results.