The Agent Skills Directory

Official Documentation Access: The skill utilizes fetch_url to retrieve content from docs.langchain.com and langchain-ai.github.io. These are official domains associated with the skill's author, used here for the intended purpose of providing accurate technical guidance.
Information Ingestion Surface: The skill processes external data fetched via the web. While this introduces a potential surface for indirect prompt injection if the source content were compromised, the use of official, vendor-controlled documentation repositories is a standard and generally safe practice.
Evidence Chain for External Data:
Ingestion points: The fetch_url tool is used in SKILL.md to read content from the documentation index and specific tutorial pages.
Boundary markers: There are no explicit delimiters or 'ignore' instructions specified in the prompt when processing the fetched content.
Capability inventory: The skill is restricted to information retrieval and does not involve subprocess execution, file system writes, or privileged commands.
Sanitization: No explicit sanitization or filtering of the retrieved text is mentioned prior to processing.

langgraph-docs