jupyter-notebook-writing

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

• Third-party content exposure detected (high risk: 0.90). The skill's reference document references/bootcamp-format.md contains an explicit example that uses LangChain's WebBaseLoader to fetch public web pages (e.g., https://lilianweng.github.io/...) and load those untrusted third‑party documents into a Milvus vector store which the agent then queries (similarity_search) to generate responses, meaning fetched web content is read and can materially influence agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The example notebook uses WebBaseLoader to fetch external documents at https://lilianweng.github.io/posts/2023-06-23-agent/ and https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/ at runtime and injects that fetched content into the model context/prompt, meaning remote content can directly control prompts.