Langchain Dependencies
Set up or upgrade LangChain, LangGraph, LangSmith, and Deep Agents with correct packages, minimum versions, and Python or TypeScript environment constraints.
Overview
langchain-dependencies is an agent skill most often used in Build (also Validate, Operate) that defines required packages, versions, and environment rules for LangChain, LangGraph, LangSmith, and Deep Agents in Python an
Install
npx skills add https://github.com/langchain-ai/langchain-skills --skill langchain-dependenciesWhat is this skill?
- LangChain 1.0 LTS guidance: new projects on 1.0+, avoid 0.3 for greenfield work
- Requires langchain-core alongside other packages; conservative pinning for langchain-community
- LangGraph vs Deep Agents framed as alternative orchestration stacks, not mandatory together
- Environment table for Python 3.10+ and TypeScript/Node minimums
- Provider integrations installed à la carte to limit dependency bloat
- LangChain 1.0 is the current LTS release
- Python 3.10+ minimum for the documented Python stack
Adoption & trust: 7.4k installs on skills.sh; 782 GitHub stars; 3/3 security scanners passed (skills.sh audits); trending (+100% hot-view momentum).
What problem does it solve?
You are starting or upgrading an agent project and do not know which LangChain-family packages, versions, or runtimes fit together without conflicts.
Who is it for?
Solo builders bootstrapping agent repos or answering 'what should I pip/npm install?' before any LangGraph or LangChain coding session.
Skip if: Projects with no LangChain ecosystem components, or teams that only need runtime graph patterns without package setup—use langgraph-fundamentals instead.
When should I use this skill?
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.
What do I get? / Deliverables
You get a coherent dependency set on LangChain 1.0 LTS principles with langchain-core explicit and orchestration choice documented—ready to implement graphs or chains.
- Recommended package list with version constraints
- Framework choice note for LangGraph vs Deep Agents
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical placement is Build → integrations because dependency choices lock model providers, orchestration, and observability before agent code lands. Integrations subphase covers package matrices and install commands—not graph logic (agent-tooling) or app UI.
Where it fits
Pick LangGraph-only deps for a spike without pulling unused vector-store clients.
Add OpenAI and LangSmith packages with versions that match langchain-core 1.0.
Plan a quarterly bump while keeping langchain-community pinned conservatively.
How it compares
Version and install matrix skill, not application architecture—invoke before fundamentals when the repo is empty or broken on upgrades.
Common Questions / FAQ
Who is langchain-dependencies for?
Developers shipping with LangChain, LangGraph, LangSmith, or Deep Agents who need authoritative package and version guidance on Prism-tagged agent workflows.
When should I use langchain-dependencies?
At validate when prototyping which stack to adopt; at build on every new repo or provider add-on; at operate when planning ecosystem upgrades and pin changes.
Is langchain-dependencies safe to install?
The skill advises versions only; verify packages on PyPI/npm and review Security Audits on this page before installing community integrations in production.
Workflow Chain
Then invoke: langgraph fundamentals
SKILL.md
READMESKILL.md - Langchain Dependencies
<overview> The LangChain ecosystem is split into focused, independently-versioned packages. Understanding which packages you need — and their version constraints — prevents incompatibilities and keeps upgrades predictable. **Key principles:** - **LangChain 1.0 is the current LTS release.** Always start new projects on 1.0+. LangChain 0.3 is legacy maintenance-only — do not use it for new work. - **langchain-core** is the shared foundation: always install it explicitly alongside any other package. - **langchain-community** (Python only) does NOT follow semantic versioning; pin it conservatively. - **LangGraph vs Deep Agents:** choose one orchestration approach based on your use case — they are alternatives, not a required stack (see [Framework Choice](#framework-choice) below). - Provider integrations (model, vector store, tools) are installed separately so you only pull in what you use. </overview> --- ## Environment Requirements <environment-requirements> | Requirement | Python | TypeScript / Node | |-------------|--------|-------------------| | Runtime minimum | **Python 3.10+** | **Node.js 20+** | | LangChain | **1.0+ (LTS)** | **1.0+ (LTS)** | | LangSmith SDK | >= 0.3.0 | >= 0.3.0 | </environment-requirements> --- ## Framework Choice <framework-choice> Pick **one** agent orchestration layer. You do not need both. | Framework | When to use | Core extra package | |-----------|-------------|--------------------| | **LangGraph** | Need fine-grained graph control, custom workflows, loops, or branching | `langgraph` / `@langchain/langgraph` | | **Deep Agents** | Want batteries-included planning, memory, file context, and skills out of the box | `deepagents` (depends on LangGraph; installs it as a transitive dep) | Both sit on top of `langchain` + `langchain-core` + `langsmith`. </framework-choice> --- ## Core Packages <python-packages> ### Python — always required | Package | Role | Min version | |---------|------|-------------| | `langchain` | Agents, chains, retrieval | 1.0 | | `langchain-core` | Base types & interfaces (peer dep) | 1.0 | | `langsmith` | Tracing, evaluation, datasets | 0.3.0 | ### Python — orchestration (pick one) | Package | Use when | Min version | |---------|----------|-------------| | `langgraph` | Building custom graphs directly | 1.0 | | `deepagents` | Using the Deep Agents framework | latest | ### Python — model providers (pick the one(s) you use) | Package | Provider | |---------|----------| | `langchain-openai` | OpenAI (GPT-4o, o3, …) | | `langchain-anthropic` | Anthropic (Claude) | | `langchain-google-genai` | Google (Gemini) | | `langchain-mistralai` | Mistral | | `langchain-groq` | Groq (fast inference) | | `langchain-cohere` | Cohere | | `langchain-fireworks` | Fireworks AI | | `langchain-together` | Together AI | | `langchain-huggingface` | Hugging Face Hub | | `langchain-ollama` | Ollama (local models) | | `langchain-aws` | AWS Bedrock | | `langchain-azure-ai` | Azure AI Foundry | ### Python — common tool & retrieval packages These packages have tighter compatibility requirements — use the latest available version unless you have a specific reason not to. | Package | Adds | Notes | |---------|------|-------| | `langchain-tavily` | Tavily web search (`TavilySearch`) | Dedicated integration package; prefer latest | | `langchain-text-splitters` | Text chunking utilities | Semver, keep current | | `langchain-community` | 1000+ integrations (fallback) | **NOT semver — pin to minor series** | | `faiss-cpu` | FAISS vector store (local) | Via `langchain-community`; use latest | | `langchain-c