LangChain Dependencies Skill: Right Packages and Versions with Claude

Teaches Claude which LangChain, LangGraph, LangSmith and Deep Agents packages and versions to use, in Python and TypeScript.

Installs: 16.4K; on skills.sh
Stars: 1.3K; on GitHub
Updated: Oct 2026; last repo commit

LangChain Dependencies is a LangChain guide that Claude reads when you start a project or ask about package versions for LangChain, LangGraph, LangSmith or Deep Agents. It lists the required packages, minimum versions, runtime requirements and the mistakes that break installs.

It is a reference and bundles only SKILL.md, no scripts. It shows requirements files and package lists, and you decide whether to run any install.

Good for

Developers starting a new LangChain or LangGraph project
Teams upgrading from LangChain 0.3 to 1.0
TypeScript and Python users who hit peer dependency or import errors

Install langchain-dependencies

Claude Code

Copy this prompt.

Install the langchain-dependencies skill in Claude Code. Run this command:

/plugin marketplace add langchain-ai/langchain-skills
/plugin install langchain-skills@langchain-skills

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

langchain-dependencies stands alone: its SKILL.md does not call other skills. The plugin route installs all 21 skills in langchain-skills (plugin.json points at config/skills/ and declares no hooks or MCP servers). The README marks the project as early development. Pick one route, not both.

From the project README. Source.

What's in the skill

When it loads

INVOKE THIS SKILL when setting up a new project or when asked about package versions, installation, or dependency management for LangChain, LangGraph, LangSmith, or Deep Agents. Covers required packages, minimum versions, environment requirements, versioning best practices, and common community tool packages for both Python and TypeScript.

Sections in SKILL.md

Environment Requirements
Framework Choice
Core Packages
Python — always required
Python — orchestration (pick one)
Python — model providers (pick the one(s) you use)
Python — common tool & retrieval packages
TypeScript — always required
TypeScript — orchestration (pick one)
TypeScript — model providers (pick the one(s) you use)
TypeScript — common tool & retrieval packages
Minimal Project Templates
Versioning Policy & Upgrade Strategy
Environment Variables
Common Mistakes

Key facts

Author: LangChain
License: MIT
Installs on skills.sh: 16,356 (Nov 2, 2026)
GitHub stars: 1,320 for the whole langchain-ai/langchain-skills repo (Nov 2, 2026)
Last commit: Oct 29, 2026
What is inside: 1 file: SKILL.md. No scripts or other files.
Starts: Claude loads it when a task matches its description, which begins "INVOKE THIS SKILL"
Repo status: The README marks the project as early development; APIs and skill content may change

What it includes

Python 3.10+, Node.js 20+, LangChain 1.0+ and LangSmith 0.3.0 or newer
The runtime and package minimums for a new project.
Environment requirements
Pick LangGraph or Deep Agents, not both
LangGraph gives graph-level control. Deep Agents gives planning, memory, file context and skills out of the box and pulls in LangGraph itself.
Framework choice
Required, orchestration, model provider and tool packages for Python and TypeScript
Tables per language of which packages you always need and which you choose by provider.
Core packages
Six dependency lists: LangGraph, Deep Agents and with-tools, each in Python and TypeScript
Starter lists you can copy, such as a requirements.txt for a LangGraph project.
Minimal project templates
Which packages follow semver and how to pin them
langchain and langgraph allow minor updates within 1.x. langchain-community is not semver, so pin the exact minor. deepagents should be pinned in production.
Versioning policy
API keys for LangSmith, model providers, Tavily and Pinecone
Lists the variable names, such as LANGSMITH_API_KEY and OPENAI_API_KEY. Set only the ones you use.
Environment variables
Seven wrong-versus-right examples, from legacy 0.3 pins to missing @langchain/core
Covers legacy version, unpinned community package, outdated tool package, deprecated community imports, missing core, old Python and old Node.
Common mistakes

From the official docs

Questions

What does the langchain-dependencies skill do?

It gives Claude LangChain's list of which packages and versions to install for LangChain, LangGraph, LangSmith and Deep Agents, plus the pinning rules and common mistakes.

How do I install it?

Use the install box at the top of this page. The langchain-skills README also documents a Claude Code plugin and npx skills for other agents.

Does it install packages for me?

The skill bundles only SKILL.md and no scripts. It tells Claude what to put in requirements or package files.

Does it need an API key?

No, not for the skill. The projects it helps set up read keys such as LANGSMITH_API_KEY or OPENAI_API_KEY from your environment, and SKILL.md says to set only the ones you use.

Is it safe to install?

It is one text file with no scripts. It contains package names and version advice, so check any install command Claude proposes before you run it.

Which Python and Node versions does it assume?

Python 3.10 or newer and Node.js 20 or newer. SKILL.md says LangChain 1.0 does not support Python 3.9 and below, and that Node below 20 is not officially supported.

Information

updated Nov 2, 2026

Repository: langchain-ai/langchain-skills
Folder: config/skills/langchain-dependencies

langchain-ai/langchain-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 3

github.com · github.com
github.com · github.com
github.com · github.com