[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C5VSCK7FZ023XFPCW7R5S2":3,"skill-same:01M6C5VSCK7FZ023XFPCW7R5S2":62},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":36,"linkOnly":37,"markdownContent":38,"markdownBody":39,"frontMatter":40,"refName":41,"commitSha":42,"skillDirectory":43,"forksCount":14,"repositoryHtmlUrl":44,"repositoryHomepageUrl":6,"ownerHtmlUrl":45,"ownerBlogUrl":46,"ownerDescription":6,"updatedAt":47,"install":48,"resources":61},"01M6C5VSCK7FZ023XFPCW7R5S2","langchain-dependencies",null,"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.","CoralShades\u002Facm-ai","CoralShades","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F220297585?v=4","01M6C5211QSJHZ8HXSDDWCR507","Coral Shades AI Agency","01M6C52161FNQDRQ0FRR8DAA9D",0,"MIT","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.continue\u002Fskills\u002Flangchain-dependencies\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35],"langchain dependencies","langchain","versioning","up","dependency","environment","dependency management","management","common","typescript","tool","new project","langgraph","package","best practices","packages","python","2026-11-10 10:12:30",false,"---\nname: langchain-dependencies\ndescription: \"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.\"\n---\n\n\u003Coverview>\nThe LangChain ecosystem is split into focused, independently-versioned packages. Understanding which packages you need — and their version constraints — prevents incompatibilities and keeps upgrades predictable.\n\n**Key principles:**\n- **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.\n- **langchain-core** is the shared foundation: always install it explicitly alongside any other package.\n- **langchain-community** (Python only) does NOT follow semantic versioning; pin it conservatively.\n- **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).\n- Provider integrations (model, vector store, tools) are installed separately so you only pull in what you use.\n\u003C\u002Foverview>\n\n---\n\n## Environment Requirements\n\n\u003Cenvironment-requirements>\n\n| Requirement | Python | TypeScript \u002F Node |\n|-------------|--------|-------------------|\n| Runtime minimum | **Python 3.10+** | **Node.js 20+** |\n| LangChain | **1.0+ (LTS)** | **1.0+ (LTS)** |\n| LangSmith SDK | >= 0.3.0 | >= 0.3.0 |\n\n\u003C\u002Fenvironment-requirements>\n\n---\n\n## Framework Choice\n\n\u003Cframework-choice>\nPick **one** agent orchestration layer. You do not need both.\n\n| Framework | When to use | Core extra package |\n|-----------|-------------|--------------------|\n| **LangGraph** | Need fine-grained graph control, custom workflows, loops, or branching | `langgraph` \u002F `@langchain\u002Flanggraph` |\n| **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) |\n\nBoth sit on top of `langchain` + `langchain-core` + `langsmith`.\n\u003C\u002Fframework-choice>\n\n---\n\n## Core Packages\n\n\u003Cpython-packages>\n\n### Python — always required\n\n| Package | Role | Min version |\n|---------|------|-------------|\n| `langchain` | Agents, chains, retrieval | 1.0 |\n| `langchain-core` | Base types & interfaces (peer dep) | 1.0 |\n| `langsmith` | Tracing, evaluation, datasets | 0.3.0 |\n\n### Python — orchestration (pick one)\n\n| Package | Use when | Min version |\n|---------|----------|-------------|\n| `langgraph` | Building custom graphs directly | 1.0 |\n| `deepagents` | Using the Deep Agents framework | latest |\n\n### Python — model providers (pick the one(s) you use)\n\n| Package | Provider |\n|---------|----------|\n| `langchain-openai` | OpenAI (GPT-4o, o3, …) |\n| `langchain-anthropic` | Anthropic (Claude) |\n| `langchain-google-genai` | Google (Gemini) |\n| `langchain-mistralai` | Mistral |\n| `langchain-groq` | Groq (fast inference) |\n| `langchain-cohere` | Cohere |\n| `langchain-fireworks` | Fireworks AI |\n| `langchain-together` | Together AI |\n| `langchain-huggingface` | Hugging Face Hub |\n| `langchain-ollama` | Ollama (local models) |\n| `langchain-aws` | AWS Bedrock |\n| `langchain-azure-ai` | Azure AI Foundry |\n\n### Python — common tool & retrieval packages\n\nThese packages have tighter compatibility requirements — use the latest available version unless you have a specific reason not to.\n\n| Package | Adds | Notes |\n|---------|------|-------|\n| `langchain-tavily` | Tavily web search (`TavilySearch`) | Dedicated integration package; prefer latest |\n| `langchain-text-splitters` | Text chunking utilities | Semver, keep current |\n| `langchain-community` | 1000+ integrations (fallback) | **NOT semver — pin to minor series** |\n| `faiss-cpu` | FAISS vector store (local) | Via `langchain-community`; use latest |\n| `langchain-chroma` | Chroma vector store | Dedicated integration package; prefer latest |\n| `langchain-pinecone` | Pinecone vector store | Dedicated integration package; prefer latest |\n| `langchain-qdrant` | Qdrant vector store | Dedicated integration package; prefer latest |\n| `langchain-weaviate` | Weaviate vector store | Dedicated integration package; prefer latest |\n| `langsmith[pytest]` | pytest plugin for LangSmith | Requires langsmith >= 0.3.4 |\n\n> **langchain-community stability note:** This package is NOT on semantic versioning. Minor releases can contain breaking changes. Prefer dedicated integration packages (e.g. `langchain-chroma`, `langchain-tavily`) when they exist — they are independently versioned and more stable.\n\n\u003C\u002Fpython-packages>\n\n\u003Ctypescript-packages>\n\n### TypeScript — always required\n\n| Package | Role | Min version |\n|---------|------|-------------|\n| `@langchain\u002Fcore` | Base types & interfaces (peer dep) | 1.0 |\n| `langchain` | Agents, chains, retrieval | 1.0 |\n| `langsmith` | Tracing, evaluation, datasets | 0.3.0 |\n\n### TypeScript — orchestration (pick one)\n\n| Package | Use when | Min version |\n|---------|----------|-------------|\n| `@langchain\u002Flanggraph` | Building custom graphs directly | 1.0 |\n| `deepagents` | Using the Deep Agents framework | latest |\n\n### TypeScript — model providers (pick the one(s) you use)\n\n| Package | Provider |\n|---------|----------|\n| `@langchain\u002Fopenai` | OpenAI (GPT-4o, o3, …) |\n| `@langchain\u002Fanthropic` | Anthropic (Claude) |\n| `@langchain\u002Fgoogle-genai` | Google (Gemini) |\n| `@langchain\u002Fmistralai` | Mistral |\n| `@langchain\u002Fgroq` | Groq (fast inference) |\n| `@langchain\u002Fcohere` | Cohere |\n| `@langchain\u002Faws` | AWS Bedrock |\n| `@langchain\u002Fazure-openai` | Azure OpenAI |\n| `@langchain\u002Follama` | Ollama (local models) |\n\n### TypeScript — common tool & retrieval packages\n\n| Package | Adds | Notes |\n|---------|------|-------|\n| `@langchain\u002Ftavily` | Tavily web search (`TavilySearch`) | Dedicated integration package; prefer latest |\n| `@langchain\u002Fcommunity` | Broad set of community integrations | Use sparingly; prefer dedicated packages |\n| `@langchain\u002Fpinecone` | Pinecone vector store | Dedicated integration package; prefer latest |\n| `@langchain\u002Fqdrant` | Qdrant vector store | Dedicated integration package; prefer latest |\n| `@langchain\u002Fweaviate` | Weaviate vector store | Dedicated integration package; prefer latest |\n\n> **`@langchain\u002Fcore` must be installed explicitly** in yarn workspaces and monorepos — it is a peer dependency and will not always be hoisted automatically.\n\n\u003C\u002Ftypescript-packages>\n\n---\n\n## Minimal Project Templates\n\n\u003Cex-langgraph-python>\n\u003Cpython>\nMinimal dependency set for a LangGraph project (provider-agnostic).\n```\n# requirements.txt\nlangchain>=1.0,\u003C2.0\nlangchain-core>=1.0,\u003C2.0\nlanggraph>=1.0,\u003C2.0\nlangsmith>=0.3.0\n\n# Add your model provider, e.g.:\n# langchain-openai\n# langchain-anthropic\n# langchain-google-genai\n```\n\u003C\u002Fpython>\n\u003C\u002Fex-langgraph-python>\n\n\u003Cex-langgraph-typescript>\n\u003Ctypescript>\nMinimal package.json dependencies for a LangGraph project (provider-agnostic).\n```json\n{\n  \"dependencies\": {\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"langchain\": \"^1.0.0\",\n    \"@langchain\u002Flanggraph\": \"^1.0.0\",\n    \"langsmith\": \"^0.3.0\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Fex-langgraph-typescript>\n\n\u003Cex-deepagents-python>\n\u003Cpython>\nMinimal dependency set for a Deep Agents project (provider-agnostic).\n```\n# requirements.txt\ndeepagents            # bundles langgraph internally\nlangchain>=1.0,\u003C2.0\nlangchain-core>=1.0,\u003C2.0\nlangsmith>=0.3.0\n\n# Add your model provider, e.g.:\n# langchain-anthropic\n# langchain-openai\n```\n\u003C\u002Fpython>\n\u003C\u002Fex-deepagents-python>\n\n\u003Cex-deepagents-typescript>\n\u003Ctypescript>\nMinimal package.json dependencies for a Deep Agents project (provider-agnostic).\n```json\n{\n  \"dependencies\": {\n    \"deepagents\": \"latest\",\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"langchain\": \"^1.0.0\",\n    \"langsmith\": \"^0.3.0\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Fex-deepagents-typescript>\n\n\u003Cex-with-tools-python>\n\u003Cpython>\nAdding Tavily search and a vector store to a LangGraph project.\n```\n# requirements.txt\nlangchain>=1.0,\u003C2.0\nlangchain-core>=1.0,\u003C2.0\nlanggraph>=1.0,\u003C2.0\nlangsmith>=0.3.0\n\n# Web search\nlangchain-tavily          # use latest; partner package, semver\n\n# Vector store — pick one:\nlangchain-chroma          # use latest; partner package, semver\n# langchain-pinecone      # use latest; partner package, semver\n# langchain-qdrant        # use latest; partner package, semver\n\n# Text processing\nlangchain-text-splitters  # use latest; semver\n\n# Your model provider:\n# langchain-openai \u002F langchain-anthropic \u002F etc.\n```\n\u003C\u002Fpython>\n\u003C\u002Fex-with-tools-python>\n\n\u003Cex-with-tools-typescript>\n\u003Ctypescript>\nAdding Tavily search and a vector store to a LangGraph project.\n```json\n{\n  \"dependencies\": {\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"langchain\": \"^1.0.0\",\n    \"@langchain\u002Flanggraph\": \"^1.0.0\",\n    \"langsmith\": \"^0.3.0\",\n    \"@langchain\u002Ftavily\": \"latest\",\n    \"@langchain\u002Fpinecone\": \"latest\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Fex-with-tools-typescript>\n\n---\n\n## Versioning Policy & Upgrade Strategy\n\n\u003Cversioning-policy>\n\n| Package group | Versioning | Safe upgrade strategy |\n|---------------|------------|-----------------------|\n| `langchain`, `langchain-core` | Strict semver (1.0 LTS) | Allow minor: `>=1.0,\u003C2.0` |\n| `langgraph` \u002F `@langchain\u002Flanggraph` | Strict semver (v1 LTS) | Allow minor: `>=1.0,\u003C2.0` |\n| `langsmith` | Strict semver | Allow minor: `>=0.3.0` |\n| Dedicated integration packages (e.g. `langchain-tavily`, `langchain-chroma`) | Independently versioned | Allow minor updates; use latest |\n| `langchain-community` | **NOT semver** | Pin exact minor: `>=0.4.0,\u003C0.5.0` |\n| `deepagents` | Follow project releases | Pin to tested version in production |\n\n**Breaking changes only happen in major versions** (1.x → 2.x) for all semver-compliant packages. Deprecated features remain functional across the entire 1.x series with warnings.\n\n**Prefer dedicated integration packages over langchain-community.** When a dedicated package exists (e.g. `langchain-chroma` instead of `langchain-community`'s Chroma integration), use it — dedicated packages are independently versioned and better tested.\n\n**Community tool packages (Tavily, vector stores, etc.) should be kept at latest** unless your project requires a locked environment. These packages frequently release compatibility fixes alongside LangChain\u002FLangGraph updates.\n\n\u003C\u002Fversioning-policy>\n\n---\n\n## Environment Variables\n\n\u003Cenvironment-variables>\nAll keys are read from the environment at runtime. Set only the keys for services you actually use.\n\n```bash\n# LangSmith (always recommended for observability)\nLANGSMITH_API_KEY=\u003Cyour-key>\nLANGSMITH_PROJECT=\u003Cproject-name>   # optional, defaults to \"default\"\n\n# Model provider — set the one(s) you use\nOPENAI_API_KEY=\u003Cyour-key>\nANTHROPIC_API_KEY=\u003Cyour-key>\nGOOGLE_API_KEY=\u003Cyour-key>\nMISTRAL_API_KEY=\u003Cyour-key>\nGROQ_API_KEY=\u003Cyour-key>\nCOHERE_API_KEY=\u003Cyour-key>\nFIREWORKS_API_KEY=\u003Cyour-key>\nTOGETHER_API_KEY=\u003Cyour-key>\nHUGGINGFACEHUB_API_TOKEN=\u003Cyour-key>\n\n# Common tool\u002Fretrieval services\nTAVILY_API_KEY=\u003Cyour-key>          # for Tavily search\nPINECONE_API_KEY=\u003Cyour-key>        # for Pinecone\n```\n\u003C\u002Fenvironment-variables>\n\n---\n\n## Common Mistakes\n\n\u003Cfix-legacy-version>\nNever start a new project on LangChain 0.3. It is maintenance-only until December 2026.\n```\n# WRONG: legacy, no new features, security patches only\nlangchain>=0.3,\u003C0.4\n\n# CORRECT: LangChain 1.0 LTS\nlangchain>=1.0,\u003C2.0\n```\n\u003C\u002Ffix-legacy-version>\n\n\u003Cfix-community-unpinned>\n`langchain-community` can break on minor version bumps — it does not follow semver.\n```\n# WRONG: allows minor-version updates that may be breaking\nlangchain-community>=0.4\n\n# CORRECT: pin to exact minor series\nlangchain-community>=0.4.0,\u003C0.5.0\n```\nAlso consider switching to the equivalent dedicated integration package if one exists (e.g. `langchain-chroma` instead of the community Chroma integration).\n\u003C\u002Ffix-community-unpinned>\n\n\u003Cfix-community-tool-outdated>\nCommunity tool packages like `langchain-tavily` and vector store integrations release compatibility fixes alongside LangChain updates. Using an old pinned version can cause import errors or broken tool schemas.\n```\n# RISKY: old pin may be incompatible with LangChain 1.0\nlangchain-tavily==0.0.1\n\n# BETTER: allow latest within the current major\nlangchain-tavily>=0.1\n```\n\u003C\u002Ffix-community-tool-outdated>\n\n\u003Cfix-community-import-deprecated>\nMany tools that used to live in `langchain-community` now have dedicated packages with updated import paths. Always prefer the dedicated package import.\n\n```python\n# WRONG — deprecated community import path\nfrom langchain_community.tools.tavily_search import TavilySearchResults\nfrom langchain_community.tools import WikipediaQueryRun\nfrom langchain_community.vectorstores import Chroma\nfrom langchain_community.vectorstores import Pinecone\n\n# CORRECT — use dedicated package imports\nfrom langchain_tavily import TavilySearch                  # pip: langchain-tavily (TavilySearchResults is deprecated)\nfrom langchain_community.tools import WikipediaQueryRun  # no dedicated pkg yet\nfrom langchain_chroma import Chroma                       # pip: langchain-chroma\nfrom langchain_pinecone import PineconeVectorStore        # pip: langchain-pinecone\n```\n\nTo find the current canonical import for any integration, search the integrations directory:\nhttps:\u002F\u002Fpython.langchain.com\u002Fdocs\u002Fintegrations\u002Ftools\u002F\n\nEach entry shows the correct package and import path. If a dedicated package exists, use it — the community path may still work but is considered legacy.\n\u003C\u002Ffix-community-import-deprecated>\n\n\u003Cfix-core-not-installed>\n\u003Ctypescript>\n`@langchain\u002Fcore` is a peer dependency — it must be in your package.json, especially in monorepos.\n```json\n\u002F\u002F WRONG: missing @langchain\u002Fcore (breaks in yarn workspaces \u002F strict hoisting)\n{\n  \"dependencies\": {\n    \"@langchain\u002Flanggraph\": \"^1.0.0\"\n  }\n}\n\n\u002F\u002F CORRECT: always list @langchain\u002Fcore explicitly\n{\n  \"dependencies\": {\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"@langchain\u002Flanggraph\": \"^1.0.0\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Ffix-core-not-installed>\n\n\u003Cfix-python-version>\n\u003Cpython>\nPython 3.9 and below are not supported by LangChain 1.0.\n```python\n# Verify before installing\nimport sys\nassert sys.version_info >= (3, 10), \"Python 3.10+ required for LangChain 1.0\"\n```\n\u003C\u002Fpython>\n\u003C\u002Ffix-python-version>\n\n\u003Cfix-node-version>\n\u003Ctypescript>\nNode.js below 20 is not officially supported.\n```bash\n# Verify before installing\nnode --version   # must be v20.x or higher\n```\n\u003C\u002Ftypescript>\n\u003C\u002Ffix-node-version>\n","\u003Coverview>\nThe LangChain ecosystem is split into focused, independently-versioned packages. Understanding which packages you need — and their version constraints — prevents incompatibilities and keeps upgrades predictable.\n\n**Key principles:**\n- **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.\n- **langchain-core** is the shared foundation: always install it explicitly alongside any other package.\n- **langchain-community** (Python only) does NOT follow semantic versioning; pin it conservatively.\n- **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).\n- Provider integrations (model, vector store, tools) are installed separately so you only pull in what you use.\n\u003C\u002Foverview>\n\n---\n\n## Environment Requirements\n\n\u003Cenvironment-requirements>\n\n| Requirement | Python | TypeScript \u002F Node |\n|-------------|--------|-------------------|\n| Runtime minimum | **Python 3.10+** | **Node.js 20+** |\n| LangChain | **1.0+ (LTS)** | **1.0+ (LTS)** |\n| LangSmith SDK | >= 0.3.0 | >= 0.3.0 |\n\n\u003C\u002Fenvironment-requirements>\n\n---\n\n## Framework Choice\n\n\u003Cframework-choice>\nPick **one** agent orchestration layer. You do not need both.\n\n| Framework | When to use | Core extra package |\n|-----------|-------------|--------------------|\n| **LangGraph** | Need fine-grained graph control, custom workflows, loops, or branching | `langgraph` \u002F `@langchain\u002Flanggraph` |\n| **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) |\n\nBoth sit on top of `langchain` + `langchain-core` + `langsmith`.\n\u003C\u002Fframework-choice>\n\n---\n\n## Core Packages\n\n\u003Cpython-packages>\n\n### Python — always required\n\n| Package | Role | Min version |\n|---------|------|-------------|\n| `langchain` | Agents, chains, retrieval | 1.0 |\n| `langchain-core` | Base types & interfaces (peer dep) | 1.0 |\n| `langsmith` | Tracing, evaluation, datasets | 0.3.0 |\n\n### Python — orchestration (pick one)\n\n| Package | Use when | Min version |\n|---------|----------|-------------|\n| `langgraph` | Building custom graphs directly | 1.0 |\n| `deepagents` | Using the Deep Agents framework | latest |\n\n### Python — model providers (pick the one(s) you use)\n\n| Package | Provider |\n|---------|----------|\n| `langchain-openai` | OpenAI (GPT-4o, o3, …) |\n| `langchain-anthropic` | Anthropic (Claude) |\n| `langchain-google-genai` | Google (Gemini) |\n| `langchain-mistralai` | Mistral |\n| `langchain-groq` | Groq (fast inference) |\n| `langchain-cohere` | Cohere |\n| `langchain-fireworks` | Fireworks AI |\n| `langchain-together` | Together AI |\n| `langchain-huggingface` | Hugging Face Hub |\n| `langchain-ollama` | Ollama (local models) |\n| `langchain-aws` | AWS Bedrock |\n| `langchain-azure-ai` | Azure AI Foundry |\n\n### Python — common tool & retrieval packages\n\nThese packages have tighter compatibility requirements — use the latest available version unless you have a specific reason not to.\n\n| Package | Adds | Notes |\n|---------|------|-------|\n| `langchain-tavily` | Tavily web search (`TavilySearch`) | Dedicated integration package; prefer latest |\n| `langchain-text-splitters` | Text chunking utilities | Semver, keep current |\n| `langchain-community` | 1000+ integrations (fallback) | **NOT semver — pin to minor series** |\n| `faiss-cpu` | FAISS vector store (local) | Via `langchain-community`; use latest |\n| `langchain-chroma` | Chroma vector store | Dedicated integration package; prefer latest |\n| `langchain-pinecone` | Pinecone vector store | Dedicated integration package; prefer latest |\n| `langchain-qdrant` | Qdrant vector store | Dedicated integration package; prefer latest |\n| `langchain-weaviate` | Weaviate vector store | Dedicated integration package; prefer latest |\n| `langsmith[pytest]` | pytest plugin for LangSmith | Requires langsmith >= 0.3.4 |\n\n> **langchain-community stability note:** This package is NOT on semantic versioning. Minor releases can contain breaking changes. Prefer dedicated integration packages (e.g. `langchain-chroma`, `langchain-tavily`) when they exist — they are independently versioned and more stable.\n\n\u003C\u002Fpython-packages>\n\n\u003Ctypescript-packages>\n\n### TypeScript — always required\n\n| Package | Role | Min version |\n|---------|------|-------------|\n| `@langchain\u002Fcore` | Base types & interfaces (peer dep) | 1.0 |\n| `langchain` | Agents, chains, retrieval | 1.0 |\n| `langsmith` | Tracing, evaluation, datasets | 0.3.0 |\n\n### TypeScript — orchestration (pick one)\n\n| Package | Use when | Min version |\n|---------|----------|-------------|\n| `@langchain\u002Flanggraph` | Building custom graphs directly | 1.0 |\n| `deepagents` | Using the Deep Agents framework | latest |\n\n### TypeScript — model providers (pick the one(s) you use)\n\n| Package | Provider |\n|---------|----------|\n| `@langchain\u002Fopenai` | OpenAI (GPT-4o, o3, …) |\n| `@langchain\u002Fanthropic` | Anthropic (Claude) |\n| `@langchain\u002Fgoogle-genai` | Google (Gemini) |\n| `@langchain\u002Fmistralai` | Mistral |\n| `@langchain\u002Fgroq` | Groq (fast inference) |\n| `@langchain\u002Fcohere` | Cohere |\n| `@langchain\u002Faws` | AWS Bedrock |\n| `@langchain\u002Fazure-openai` | Azure OpenAI |\n| `@langchain\u002Follama` | Ollama (local models) |\n\n### TypeScript — common tool & retrieval packages\n\n| Package | Adds | Notes |\n|---------|------|-------|\n| `@langchain\u002Ftavily` | Tavily web search (`TavilySearch`) | Dedicated integration package; prefer latest |\n| `@langchain\u002Fcommunity` | Broad set of community integrations | Use sparingly; prefer dedicated packages |\n| `@langchain\u002Fpinecone` | Pinecone vector store | Dedicated integration package; prefer latest |\n| `@langchain\u002Fqdrant` | Qdrant vector store | Dedicated integration package; prefer latest |\n| `@langchain\u002Fweaviate` | Weaviate vector store | Dedicated integration package; prefer latest |\n\n> **`@langchain\u002Fcore` must be installed explicitly** in yarn workspaces and monorepos — it is a peer dependency and will not always be hoisted automatically.\n\n\u003C\u002Ftypescript-packages>\n\n---\n\n## Minimal Project Templates\n\n\u003Cex-langgraph-python>\n\u003Cpython>\nMinimal dependency set for a LangGraph project (provider-agnostic).\n```\n# requirements.txt\nlangchain>=1.0,\u003C2.0\nlangchain-core>=1.0,\u003C2.0\nlanggraph>=1.0,\u003C2.0\nlangsmith>=0.3.0\n\n# Add your model provider, e.g.:\n# langchain-openai\n# langchain-anthropic\n# langchain-google-genai\n```\n\u003C\u002Fpython>\n\u003C\u002Fex-langgraph-python>\n\n\u003Cex-langgraph-typescript>\n\u003Ctypescript>\nMinimal package.json dependencies for a LangGraph project (provider-agnostic).\n```json\n{\n  \"dependencies\": {\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"langchain\": \"^1.0.0\",\n    \"@langchain\u002Flanggraph\": \"^1.0.0\",\n    \"langsmith\": \"^0.3.0\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Fex-langgraph-typescript>\n\n\u003Cex-deepagents-python>\n\u003Cpython>\nMinimal dependency set for a Deep Agents project (provider-agnostic).\n```\n# requirements.txt\ndeepagents            # bundles langgraph internally\nlangchain>=1.0,\u003C2.0\nlangchain-core>=1.0,\u003C2.0\nlangsmith>=0.3.0\n\n# Add your model provider, e.g.:\n# langchain-anthropic\n# langchain-openai\n```\n\u003C\u002Fpython>\n\u003C\u002Fex-deepagents-python>\n\n\u003Cex-deepagents-typescript>\n\u003Ctypescript>\nMinimal package.json dependencies for a Deep Agents project (provider-agnostic).\n```json\n{\n  \"dependencies\": {\n    \"deepagents\": \"latest\",\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"langchain\": \"^1.0.0\",\n    \"langsmith\": \"^0.3.0\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Fex-deepagents-typescript>\n\n\u003Cex-with-tools-python>\n\u003Cpython>\nAdding Tavily search and a vector store to a LangGraph project.\n```\n# requirements.txt\nlangchain>=1.0,\u003C2.0\nlangchain-core>=1.0,\u003C2.0\nlanggraph>=1.0,\u003C2.0\nlangsmith>=0.3.0\n\n# Web search\nlangchain-tavily          # use latest; partner package, semver\n\n# Vector store — pick one:\nlangchain-chroma          # use latest; partner package, semver\n# langchain-pinecone      # use latest; partner package, semver\n# langchain-qdrant        # use latest; partner package, semver\n\n# Text processing\nlangchain-text-splitters  # use latest; semver\n\n# Your model provider:\n# langchain-openai \u002F langchain-anthropic \u002F etc.\n```\n\u003C\u002Fpython>\n\u003C\u002Fex-with-tools-python>\n\n\u003Cex-with-tools-typescript>\n\u003Ctypescript>\nAdding Tavily search and a vector store to a LangGraph project.\n```json\n{\n  \"dependencies\": {\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"langchain\": \"^1.0.0\",\n    \"@langchain\u002Flanggraph\": \"^1.0.0\",\n    \"langsmith\": \"^0.3.0\",\n    \"@langchain\u002Ftavily\": \"latest\",\n    \"@langchain\u002Fpinecone\": \"latest\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Fex-with-tools-typescript>\n\n---\n\n## Versioning Policy & Upgrade Strategy\n\n\u003Cversioning-policy>\n\n| Package group | Versioning | Safe upgrade strategy |\n|---------------|------------|-----------------------|\n| `langchain`, `langchain-core` | Strict semver (1.0 LTS) | Allow minor: `>=1.0,\u003C2.0` |\n| `langgraph` \u002F `@langchain\u002Flanggraph` | Strict semver (v1 LTS) | Allow minor: `>=1.0,\u003C2.0` |\n| `langsmith` | Strict semver | Allow minor: `>=0.3.0` |\n| Dedicated integration packages (e.g. `langchain-tavily`, `langchain-chroma`) | Independently versioned | Allow minor updates; use latest |\n| `langchain-community` | **NOT semver** | Pin exact minor: `>=0.4.0,\u003C0.5.0` |\n| `deepagents` | Follow project releases | Pin to tested version in production |\n\n**Breaking changes only happen in major versions** (1.x → 2.x) for all semver-compliant packages. Deprecated features remain functional across the entire 1.x series with warnings.\n\n**Prefer dedicated integration packages over langchain-community.** When a dedicated package exists (e.g. `langchain-chroma` instead of `langchain-community`'s Chroma integration), use it — dedicated packages are independently versioned and better tested.\n\n**Community tool packages (Tavily, vector stores, etc.) should be kept at latest** unless your project requires a locked environment. These packages frequently release compatibility fixes alongside LangChain\u002FLangGraph updates.\n\n\u003C\u002Fversioning-policy>\n\n---\n\n## Environment Variables\n\n\u003Cenvironment-variables>\nAll keys are read from the environment at runtime. Set only the keys for services you actually use.\n\n```bash\n# LangSmith (always recommended for observability)\nLANGSMITH_API_KEY=\u003Cyour-key>\nLANGSMITH_PROJECT=\u003Cproject-name>   # optional, defaults to \"default\"\n\n# Model provider — set the one(s) you use\nOPENAI_API_KEY=\u003Cyour-key>\nANTHROPIC_API_KEY=\u003Cyour-key>\nGOOGLE_API_KEY=\u003Cyour-key>\nMISTRAL_API_KEY=\u003Cyour-key>\nGROQ_API_KEY=\u003Cyour-key>\nCOHERE_API_KEY=\u003Cyour-key>\nFIREWORKS_API_KEY=\u003Cyour-key>\nTOGETHER_API_KEY=\u003Cyour-key>\nHUGGINGFACEHUB_API_TOKEN=\u003Cyour-key>\n\n# Common tool\u002Fretrieval services\nTAVILY_API_KEY=\u003Cyour-key>          # for Tavily search\nPINECONE_API_KEY=\u003Cyour-key>        # for Pinecone\n```\n\u003C\u002Fenvironment-variables>\n\n---\n\n## Common Mistakes\n\n\u003Cfix-legacy-version>\nNever start a new project on LangChain 0.3. It is maintenance-only until December 2026.\n```\n# WRONG: legacy, no new features, security patches only\nlangchain>=0.3,\u003C0.4\n\n# CORRECT: LangChain 1.0 LTS\nlangchain>=1.0,\u003C2.0\n```\n\u003C\u002Ffix-legacy-version>\n\n\u003Cfix-community-unpinned>\n`langchain-community` can break on minor version bumps — it does not follow semver.\n```\n# WRONG: allows minor-version updates that may be breaking\nlangchain-community>=0.4\n\n# CORRECT: pin to exact minor series\nlangchain-community>=0.4.0,\u003C0.5.0\n```\nAlso consider switching to the equivalent dedicated integration package if one exists (e.g. `langchain-chroma` instead of the community Chroma integration).\n\u003C\u002Ffix-community-unpinned>\n\n\u003Cfix-community-tool-outdated>\nCommunity tool packages like `langchain-tavily` and vector store integrations release compatibility fixes alongside LangChain updates. Using an old pinned version can cause import errors or broken tool schemas.\n```\n# RISKY: old pin may be incompatible with LangChain 1.0\nlangchain-tavily==0.0.1\n\n# BETTER: allow latest within the current major\nlangchain-tavily>=0.1\n```\n\u003C\u002Ffix-community-tool-outdated>\n\n\u003Cfix-community-import-deprecated>\nMany tools that used to live in `langchain-community` now have dedicated packages with updated import paths. Always prefer the dedicated package import.\n\n```python\n# WRONG — deprecated community import path\nfrom langchain_community.tools.tavily_search import TavilySearchResults\nfrom langchain_community.tools import WikipediaQueryRun\nfrom langchain_community.vectorstores import Chroma\nfrom langchain_community.vectorstores import Pinecone\n\n# CORRECT — use dedicated package imports\nfrom langchain_tavily import TavilySearch                  # pip: langchain-tavily (TavilySearchResults is deprecated)\nfrom langchain_community.tools import WikipediaQueryRun  # no dedicated pkg yet\nfrom langchain_chroma import Chroma                       # pip: langchain-chroma\nfrom langchain_pinecone import PineconeVectorStore        # pip: langchain-pinecone\n```\n\nTo find the current canonical import for any integration, search the integrations directory:\nhttps:\u002F\u002Fpython.langchain.com\u002Fdocs\u002Fintegrations\u002Ftools\u002F\n\nEach entry shows the correct package and import path. If a dedicated package exists, use it — the community path may still work but is considered legacy.\n\u003C\u002Ffix-community-import-deprecated>\n\n\u003Cfix-core-not-installed>\n\u003Ctypescript>\n`@langchain\u002Fcore` is a peer dependency — it must be in your package.json, especially in monorepos.\n```json\n\u002F\u002F WRONG: missing @langchain\u002Fcore (breaks in yarn workspaces \u002F strict hoisting)\n{\n  \"dependencies\": {\n    \"@langchain\u002Flanggraph\": \"^1.0.0\"\n  }\n}\n\n\u002F\u002F CORRECT: always list @langchain\u002Fcore explicitly\n{\n  \"dependencies\": {\n    \"@langchain\u002Fcore\": \"^1.0.0\",\n    \"@langchain\u002Flanggraph\": \"^1.0.0\"\n  }\n}\n```\n\u003C\u002Ftypescript>\n\u003C\u002Ffix-core-not-installed>\n\n\u003Cfix-python-version>\n\u003Cpython>\nPython 3.9 and below are not supported by LangChain 1.0.\n```python\n# Verify before installing\nimport sys\nassert sys.version_info >= (3, 10), \"Python 3.10+ required for LangChain 1.0\"\n```\n\u003C\u002Fpython>\n\u003C\u002Ffix-python-version>\n\n\u003Cfix-node-version>\n\u003Ctypescript>\nNode.js below 20 is not officially supported.\n```bash\n# Verify before installing\nnode --version   # must be v20.x or higher\n```\n\u003C\u002Ftypescript>\n\u003C\u002Ffix-node-version>\n",{"name":5,"description":7},"main","1ca81b0a7de199942400f965edd4d7544077ac5d",".continue\u002Fskills\u002Flangchain-dependencies","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai","https:\u002F\u002Fgithub.com\u002FCoralShades","coralshades.com.au","2026-04-18 00:45:07",{"source":8,"skillName":5,"commands":49,"prompt":59,"zipAvailable":60},[50,53,56],{"tool":51,"command":52},"npx","npx skills add CoralShades\u002Facm-ai --skill langchain-dependencies",{"tool":54,"command":55},"bunx","bunx skills add CoralShades\u002Facm-ai --skill langchain-dependencies",{"tool":57,"command":58},"pnpm","pnpm dlx skills add CoralShades\u002Facm-ai --skill langchain-dependencies","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.continue\u002Fskills\u002Flangchain-dependencies\u002FSKILL.md",true,[],[63,87,100,115,137,147],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":74,"publishedAt":36},"01M6C63HKXA890HHV47WBAX0TB","uncodixfy","Prevents generic AI\u002FCodex UI patterns when generating frontend code. Use this skill whenever generating HTML, CSS, React, Vue, Svelte, or any frontend UI code to enforce clean, human-designed aesthetics inspired by Linear, Raycast, Stripe, and GitHub instead of typical AI-generated UI.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Funcodixfy\u002FSKILL.md",[69],{"categoryNo":70,"categoryName":71,"taxonomyType":72,"taxonomyTypeDesc":73},"01JBSEEDCAT000000000000009","前端开发","SCENARIO","使用场景",[65,75,76,77,78,79,80,81,82,83,84,85,86],"github","frontend","css","codex","react","code","linear","ui","vue","generated","use","html",{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":91,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":36},"01M6C63FJ2GEBBGZ5XSMV40KXB","runpodctl","Runpod CLI to manage your GPU workloads.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Frunpodctl\u002FSKILL.md",[],[89,95,96,97,98,99],"macos","cli","linux","gpu","Bash",{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":36},"01M6C63DPAV46TFVQ01SFYPJ7X","langgraph-persistence","INVOKE THIS SKILL when your LangGraph needs to persist state, remember conversations, travel through history, or configure subgraph checkpointer scoping. Covers checkpointers, thread_id, time travel, Store, and subgraph persistence modes.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flanggraph-persistence\u002FSKILL.md",[],[107,31,108,109,110,111,112,113,114],"langgraph persistence","persistence","travel","history","scoping","remember","state","configure",{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":126,"publishedAt":36},"01M6C63BNS9CQEBTB8RNSSBFPA","langgraph-human-in-the-loop","INVOKE THIS SKILL when implementing human-in-the-loop patterns, pausing for approval, or handling errors in LangGraph. Covers interrupt(), Command(resume=...), approval\u002Fvalidation workflows, and the 4-tier error handling strategy.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flanggraph-human-in-the-loop\u002FSKILL.md",[121],{"categoryNo":122,"categoryName":123,"taxonomyType":124,"taxonomyTypeDesc":125},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[127,128,31,129,130,131,132,133,134,135,136],"langgraph human in the loop","loop","workflows","error handling","validation","human-in-the-loop","command","strategy","approval","resume",{"skillNo":138,"name":139,"nameZh":6,"description":140,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":141,"categories":142,"keywords":143,"publishedAt":36},"01M6C639GVYZFX3NFPV5FBDR4F","langgraph-fundamentals","INVOKE THIS SKILL when writing ANY LangGraph code. Covers StateGraph, state schemas, nodes, edges, Command, Send, invoke, streaming, and error handling.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flanggraph-fundamentals\u002FSKILL.md",[],[144,31,130,113,145,146,133,80],"langgraph fundamentals","streaming","writing",{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":36},"01M6C6375N34YX2GZ3SD3R7XR2","langchain-rag","INVOKE THIS SKILL when building ANY retrieval-augmented generation (RAG) system. Covers document loaders, RecursiveCharacterTextSplitter, embeddings (OpenAI), and vector stores (Chroma, FAISS, Pinecone).","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flangchain-rag\u002FSKILL.md",[],[154,155,20,156,157,158,159,160,161,162],"rag","langchain rag","embeddings","vector","openai","chroma","faiss","vector stores","system"]