Source record

Published by LangChain Blog (T1.5 source). The original is at https://www.langchain.com/blog/revamping-skills-in-deep-agents.

Pipeline notes

The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.

LangChain's Deep Agents now supports binding tools to skills, pinning skills at runtime, and reloading skills mid-thread. Skills use a SKILL.md file with optional scripts/, references/, and assets/ directories, loaded progressively so agents only read full instructions when a task requires them.

Concrete skill file structure and runtime behaviors for scaling skill libraries in Deep Agents, grounded in progressive disclosure for context efficiency.

Cited by

No citations on record.