whipflow

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill auto-fills "ask" inputs from the user's context and embeds those variables into the generated .whip source which it then must include verbatim when calling whipflow_run(source=...), so any API keys/passwords present in context would be output directly by the LLM.

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

• Third-party content exposure detected (high risk: 0.90). The skill's DSL and required workflows explicitly instruct agents to fetch and process web content (e.g., agent researcher with tools ["web_search","read"] in references/examples.md and the "choice" example that does "抓取网页 {input} 的内容"), meaning it ingests untrusted public webpages which can influence subsequent agent actions.