The Agent Skills Directory

Indirect Prompt Injection (LOW): The skill ingests untrusted data from external URLs (langchain.com) via the fetch_url tool. While this content could theoretically contain instructions to mislead the agent, the skill lacks dangerous capabilities such as file-system access or shell execution, which significantly limits the potential impact.
Data Exposure & Exfiltration (LOW): The skill performs network operations to docs.langchain.com. Although this domain is not on the explicit whitelist of trusted infrastructure, it is a known reputable documentation source and the skill does not attempt to access or transmit sensitive local information.

langgraph-docs