[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-langchain-ecosystem-primer":3,"mdc--ms9nj6-key":37,"related-org-langchain-ecosystem-primer":1045,"related-repo-langchain-ecosystem-primer":1222},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":27,"repoUrl":28,"updatedAt":29,"license":30,"forks":31,"topics":32,"repo":33,"sourceUrl":35,"mdContent":36},"ecosystem-primer","select frameworks for LangChain and LangGraph agents","INVOKE FIRST for any LangChain \u002F LangGraph \u002F Deep Agents agent building project before consulting other skills or writing any agent code. Required starting point for up to date info on framework selection (LangChain vs LangGraph vs Deep Agents vs hybrid composition), agent patterns, install, environment setup, and which skill to load next.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"langchain","LangChain","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Flangchain.png","langchain-ai",[13,17,18,21,24],{"name":14,"slug":15,"type":16},"Architecture","architecture","tag",{"name":9,"slug":8,"type":16},{"name":19,"slug":20,"type":16},"LangGraph","langgraph",{"name":22,"slug":23,"type":16},"Agents","agents",{"name":25,"slug":26,"type":16},"AI Infrastructure","ai-infrastructure",877,"https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flangchain-skills","2026-07-24T05:42:48.348966",null,77,[],{"repoUrl":28,"stars":27,"forks":31,"topics":34,"description":30},[],"https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flangchain-skills\u002Ftree\u002FHEAD\u002Fconfig\u002Fskills\u002Fecosystem-primer","---\nname: ecosystem-primer\ndescription: \"INVOKE FIRST for any LangChain \u002F LangGraph \u002F Deep Agents agent building project before consulting other skills or writing any agent code. Required starting point for up to date info on framework selection (LangChain vs LangGraph vs Deep Agents vs hybrid composition), agent patterns, install, environment setup, and which skill to load next.\"\n---\n\n\u003Coverview>\nLangChain Inc. maintains three layered open-source tools for building agents, plus LangSmith for observability. The stack, top-down:\n\n- **Deep Agents** (top layer, *harness*) — batteries-included toolkit built on LangChain + LangGraph. Ships with planning, file management, subagent spawning, and memory out of the box.\n- **LangGraph** (middle layer, *runtime*) — low-level orchestration for durable execution, custom control flow, and stateful workflows. LangChain agents run on top of LangGraph.\n- **LangChain** (bottom layer, *framework*) — abstractions for models, tools, and the agent loop. Provider-agnostic, easiest to start with.\n- **LangSmith** (cross-cutting) — observability and evaluation platform. Framework-agnostic; always recommended alongside any of the above.\n\nHigher layers depend on lower ones, but you don't need to use lower layers directly. Deep Agents gives you LangGraph's durable execution without writing graph code. LangChain gives you models and tools without managing graph edges.\n\u003C\u002Foverview>\n\n---\n\n## Step 1 — Choose Your Tool\n\n\u003Cdecision-table>\n\nEvaluate these conditions in order and stop at the first match:\n\n1. If the task needs planning, file management across a long session, persistent memory, subagent delegation, or on-demand skills → **Deep Agents**\n2. Else, if the task needs custom control flow (deterministic loops, branching logic) → **LangGraph**\n3. Else, if it's a single-purpose agent with a fixed set of tools → **LangChain** (`create_agent` function)\n4. Else, if it's a pure model call, retrieval pipeline, or simple prompt chain with no agent loop → **LangChain** (direct model \u002F chain)\n\nThis is your **layer**. BUT you are not done: later in Step 4, you MUST load the layer-specific skill before writing any agent code.\n\n\u003C\u002Fdecision-table>\n\n---\n\n## Tool Profiles\n\n\u003Clangchain-profile>\n\n### LangChain — agent framework\n\n**Best for:**\n- Single-purpose agents with a fixed tool set\n- RAG pipelines and document Q&A\n- Model calls, prompt templates, structured output\n\n**Not ideal when:**\n- The agent needs to plan across many steps or manage large context\n- Control flow is conditional, iterative, or parallel\n- State must persist across sessions\n\nAll LangChain agents use `create_agent(model, tools=[...])`.\n\n\u003C\u002Flangchain-profile>\n\n\u003Clanggraph-profile>\n\n### LangGraph — agent runtime\n\n**Best for:**\n- Custom control flow — deterministic loops, reflection cycles, parallel fan-out\n- Complex workflows combining deterministic and agentic steps\n- Human-in-the-loop with precise interrupt and resume points\n- State that must survive failures or span long sessions\n\n**Not ideal when:**\n- You want planning, file management, and subagent delegation out of the box (use Deep Agents instead)\n- The workflow is simple enough for a straight tool loop\n\nAll LangGraph graphs use `StateGraph(State)` with explicit nodes, edges, and conditional edges.\n\n\u003C\u002Flanggraph-profile>\n\n\u003Cdeep-agents-profile>\n\n### Deep Agents — agent harness\n\n**Best for:**\n- Long-running tasks that require planning and decomposition\n- Agents that read, write, and manage files across a session\n- Delegating subtasks to specialized subagents\n- Persistent memory across sessions\n- Loading domain-specific skills on demand\n\n**Not ideal when:**\n- The task is simple enough for a single-purpose agent\n- You need precise hand-crafted control over every graph edge (use LangGraph directly)\n\nAll Deep Agents use `create_deep_agent(model, tools=[...])`.\n\n\u003C\u002Fdeep-agents-profile>\n\n---\n\n## Mixing Layers\n\n\u003Cmixing-layers>\n\nThe tools are layered, so they can be combined in the same project. Common patterns:\n\n- **Deep Agents orchestrator → LangGraph subagent** — when the main agent needs planning and memory but one subtask requires a deterministic graph.\n- **LangGraph graph wrapped as a tool or subagent** — when a specialized pipeline (e.g. RAG, reflection loop) is called by a broader agent.\n\nA compiled LangGraph graph can be registered as a named subagent inside Deep Agents — the orchestrator delegates to it via the `task` tool without knowing its internal structure. LangChain tools and retrievers work freely inside both LangGraph nodes and Deep Agents tools.\n\n\u003C\u002Fmixing-layers>\n\n---\n\n## Step 2 — Set Environment Variables\n\nAlways set these for observability. These are the current LangSmith env var names. Copy them as-is. OLDER NAMES NO LONGER WORK.\n\n\u003Cenvironment-variables>\nLANGSMITH_API_KEY=\u003Cyour-key>\nLANGSMITH_TRACING=true\nLANGSMITH_PROJECT=\u003Cproject-name>\n\u003C\u002Fenvironment-variables>\n\nModel-provider and tool-specific keys (`ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, `TAVILY_API_KEY`, etc.) depend on your stack — set them as needed.\n\n---\n\n## Step 3 — How the Docs Work\n\n\u003Cdocs>\n\nAll documentation lives at **docs.langchain.com**, organized into two top-level sections:\n\n- **OSS** — LangChain, LangGraph, Deep Agents. Python (`\u002Foss\u002Fpython\u002F`) and TypeScript (`\u002Foss\u002Fjavascript\u002F`) trees in parallel.\n- **LangSmith** — observability, evaluation, deployment, prompt engineering.\n\nEach product has its own page tree: overview → quickstart → how-to guides → reference.\n\n### Canonical landing pages\n\nStart here rather than tree-searching from root (swap `python` → `javascript` for TypeScript):\n\n- **LangChain** — `\u002Foss\u002Fpython\u002Flangchain\u002Foverview`\n- **LangGraph** — `\u002Foss\u002Fpython\u002Flanggraph\u002Foverview`\n- **Deep Agents** — `\u002Foss\u002Fpython\u002Fdeepagents\u002Foverview`\n- **LangSmith** — `\u002Flangsmith\u002Fhome` (no language split)\n\n### Accessing docs in an agent context\n\n**If the LangChain Docs MCP server is connected** (`mcp__docs-langchain__*` tools are available), query it directly:\n```\ntree \u002Foss\u002Fpython -L 2                        # explore Python structure\ntree \u002Foss\u002Fjavascript -L 2                    # parallel TypeScript structure\ncat \u002Foss\u002Fpython\u002Flangchain\u002Fquickstart.mdx     # read a specific page\nrg -il \"checkpointer\" \u002Foss\u002Fpython\u002Flanggraph\u002F # search by keyword\n```\n\n**If the MCP server is not available**, use the `llms.txt` index:\n1. Fetch `https:\u002F\u002Fdocs.langchain.com\u002Fllms.txt` — structured list of all pages with descriptions\n2. Identify the 2–4 most relevant pages for the question\n3. Fetch those pages directly for accurate, up-to-date content\n\n> Always prefer fetching live docs over relying on training-data knowledge — these libraries evolve fast and APIs change often.\n\n\u003C\u002Fdocs>\n\n---\n\n## Step 4 — Load the Right Skill Next\n\nIf the user only wants a minimal local working agent (new project, stub tool, provider key), load the matching quickstart first:\n\n- LangChain → `langchain-python-quickstart` or `langchain-typescript-quickstart`\n- LangGraph → `langgraph-python-quickstart` or `langgraph-typescript-quickstart`\n- Deep Agents → `deepagents-python-quickstart` or `deepagents-typescript-quickstart`\n\nOtherwise load the skill below that matches your layer from Step 1. This is required — the layer-specific skill carries the current API; the primer alone does not.\n\n\u003Cnext-skills>\n\n### LangChain\n\n- **`langchain-fundamentals`** — building any LangChain agent\n- **`langchain-rag`** — adding RAG \u002F vector store retrieval\n- **`langchain-middleware`** — structured output with Pydantic\n- **`langchain-dependencies`** — package versions, installs, or dependency management questions\n\n### LangGraph\n\n- **`langgraph-fundamentals`** — any LangGraph graph\n- **`langgraph-human-in-the-loop`** — human-in-the-loop or approval workflows\n- **`langgraph-persistence`** — state that must survive restarts, or cross-thread memory\n\n### Deep Agents\n\n**Always load `deep-agents-core` first.** Then, as needed:\n\n- **`deep-agents-orchestration`** — subagent delegation or orchestration\n- **`deep-agents-memory`** — cross-session persistent memory\n\n\u003C\u002Fnext-skills>\n",{"data":38,"body":39},{"name":4,"description":6},{"type":40,"children":41},"root",[42],{"type":43,"tag":44,"props":45,"children":46},"element","overview",{},[47,50,116,122,126,133,207,210,216,292,365,442,445,451,495,498,504,509,525,553,556,562,798,801,807,812,870,875],{"type":48,"value":49},"text","\nLangChain Inc. maintains three layered open-source tools for building agents, plus LangSmith for observability. The stack, top-down:\n",{"type":43,"tag":51,"props":52,"children":53},"ul",{},[54,74,90,106],{"type":43,"tag":55,"props":56,"children":57},"li",{},[58,64,66,72],{"type":43,"tag":59,"props":60,"children":61},"strong",{},[62],{"type":48,"value":63},"Deep Agents",{"type":48,"value":65}," (top layer, ",{"type":43,"tag":67,"props":68,"children":69},"em",{},[70],{"type":48,"value":71},"harness",{"type":48,"value":73},") — batteries-included toolkit built on LangChain + LangGraph. Ships with planning, file management, subagent spawning, and memory out of the box.",{"type":43,"tag":55,"props":75,"children":76},{},[77,81,83,88],{"type":43,"tag":59,"props":78,"children":79},{},[80],{"type":48,"value":19},{"type":48,"value":82}," (middle layer, ",{"type":43,"tag":67,"props":84,"children":85},{},[86],{"type":48,"value":87},"runtime",{"type":48,"value":89},") — low-level orchestration for durable execution, custom control flow, and stateful workflows. LangChain agents run on top of LangGraph.",{"type":43,"tag":55,"props":91,"children":92},{},[93,97,99,104],{"type":43,"tag":59,"props":94,"children":95},{},[96],{"type":48,"value":9},{"type":48,"value":98}," (bottom layer, ",{"type":43,"tag":67,"props":100,"children":101},{},[102],{"type":48,"value":103},"framework",{"type":48,"value":105},") — abstractions for models, tools, and the agent loop. Provider-agnostic, easiest to start with.",{"type":43,"tag":55,"props":107,"children":108},{},[109,114],{"type":43,"tag":59,"props":110,"children":111},{},[112],{"type":48,"value":113},"LangSmith",{"type":48,"value":115}," (cross-cutting) — observability and evaluation platform. Framework-agnostic; always recommended alongside any of the above.",{"type":43,"tag":117,"props":118,"children":119},"p",{},[120],{"type":48,"value":121},"Higher layers depend on lower ones, but you don't need to use lower layers directly. Deep Agents gives you LangGraph's durable execution without writing graph code. LangChain gives you models and tools without managing graph edges.\n",{"type":43,"tag":123,"props":124,"children":125},"hr",{},[],{"type":43,"tag":127,"props":128,"children":130},"h2",{"id":129},"step-1-choose-your-tool",[131],{"type":48,"value":132},"Step 1 — Choose Your Tool",{"type":43,"tag":134,"props":135,"children":136},"decision-table",{},[137,142,195],{"type":43,"tag":117,"props":138,"children":139},{},[140],{"type":48,"value":141},"Evaluate these conditions in order and stop at the first match:",{"type":43,"tag":143,"props":144,"children":145},"ol",{},[146,155,164,184],{"type":43,"tag":55,"props":147,"children":148},{},[149,151],{"type":48,"value":150},"If the task needs planning, file management across a long session, persistent memory, subagent delegation, or on-demand skills → ",{"type":43,"tag":59,"props":152,"children":153},{},[154],{"type":48,"value":63},{"type":43,"tag":55,"props":156,"children":157},{},[158,160],{"type":48,"value":159},"Else, if the task needs custom control flow (deterministic loops, branching logic) → ",{"type":43,"tag":59,"props":161,"children":162},{},[163],{"type":48,"value":19},{"type":43,"tag":55,"props":165,"children":166},{},[167,169,173,175,182],{"type":48,"value":168},"Else, if it's a single-purpose agent with a fixed set of tools → ",{"type":43,"tag":59,"props":170,"children":171},{},[172],{"type":48,"value":9},{"type":48,"value":174}," (",{"type":43,"tag":176,"props":177,"children":179},"code",{"className":178},[],[180],{"type":48,"value":181},"create_agent",{"type":48,"value":183}," function)",{"type":43,"tag":55,"props":185,"children":186},{},[187,189,193],{"type":48,"value":188},"Else, if it's a pure model call, retrieval pipeline, or simple prompt chain with no agent loop → ",{"type":43,"tag":59,"props":190,"children":191},{},[192],{"type":48,"value":9},{"type":48,"value":194}," (direct model \u002F chain)",{"type":43,"tag":117,"props":196,"children":197},{},[198,200,205],{"type":48,"value":199},"This is your ",{"type":43,"tag":59,"props":201,"children":202},{},[203],{"type":48,"value":204},"layer",{"type":48,"value":206},". BUT you are not done: later in Step 4, you MUST load the layer-specific skill before writing any agent code.",{"type":43,"tag":123,"props":208,"children":209},{},[],{"type":43,"tag":127,"props":211,"children":213},{"id":212},"tool-profiles",[214],{"type":48,"value":215},"Tool Profiles",{"type":43,"tag":217,"props":218,"children":219},"langchain-profile",{},[220,227,235,253,261,279],{"type":43,"tag":221,"props":222,"children":224},"h3",{"id":223},"langchain-agent-framework",[225],{"type":48,"value":226},"LangChain — agent framework",{"type":43,"tag":117,"props":228,"children":229},{},[230],{"type":43,"tag":59,"props":231,"children":232},{},[233],{"type":48,"value":234},"Best for:",{"type":43,"tag":51,"props":236,"children":237},{},[238,243,248],{"type":43,"tag":55,"props":239,"children":240},{},[241],{"type":48,"value":242},"Single-purpose agents with a fixed tool set",{"type":43,"tag":55,"props":244,"children":245},{},[246],{"type":48,"value":247},"RAG pipelines and document Q&A",{"type":43,"tag":55,"props":249,"children":250},{},[251],{"type":48,"value":252},"Model calls, prompt templates, structured output",{"type":43,"tag":117,"props":254,"children":255},{},[256],{"type":43,"tag":59,"props":257,"children":258},{},[259],{"type":48,"value":260},"Not ideal when:",{"type":43,"tag":51,"props":262,"children":263},{},[264,269,274],{"type":43,"tag":55,"props":265,"children":266},{},[267],{"type":48,"value":268},"The agent needs to plan across many steps or manage large context",{"type":43,"tag":55,"props":270,"children":271},{},[272],{"type":48,"value":273},"Control flow is conditional, iterative, or parallel",{"type":43,"tag":55,"props":275,"children":276},{},[277],{"type":48,"value":278},"State must persist across sessions",{"type":43,"tag":117,"props":280,"children":281},{},[282,284,290],{"type":48,"value":283},"All LangChain agents use ",{"type":43,"tag":176,"props":285,"children":287},{"className":286},[],[288],{"type":48,"value":289},"create_agent(model, tools=[...])",{"type":48,"value":291},".",{"type":43,"tag":293,"props":294,"children":295},"langgraph-profile",{},[296,302,309,332,339,352],{"type":43,"tag":221,"props":297,"children":299},{"id":298},"langgraph-agent-runtime",[300],{"type":48,"value":301},"LangGraph — agent runtime",{"type":43,"tag":117,"props":303,"children":304},{},[305],{"type":43,"tag":59,"props":306,"children":307},{},[308],{"type":48,"value":234},{"type":43,"tag":51,"props":310,"children":311},{},[312,317,322,327],{"type":43,"tag":55,"props":313,"children":314},{},[315],{"type":48,"value":316},"Custom control flow — deterministic loops, reflection cycles, parallel fan-out",{"type":43,"tag":55,"props":318,"children":319},{},[320],{"type":48,"value":321},"Complex workflows combining deterministic and agentic steps",{"type":43,"tag":55,"props":323,"children":324},{},[325],{"type":48,"value":326},"Human-in-the-loop with precise interrupt and resume points",{"type":43,"tag":55,"props":328,"children":329},{},[330],{"type":48,"value":331},"State that must survive failures or span long sessions",{"type":43,"tag":117,"props":333,"children":334},{},[335],{"type":43,"tag":59,"props":336,"children":337},{},[338],{"type":48,"value":260},{"type":43,"tag":51,"props":340,"children":341},{},[342,347],{"type":43,"tag":55,"props":343,"children":344},{},[345],{"type":48,"value":346},"You want planning, file management, and subagent delegation out of the box (use Deep Agents instead)",{"type":43,"tag":55,"props":348,"children":349},{},[350],{"type":48,"value":351},"The workflow is simple enough for a straight tool loop",{"type":43,"tag":117,"props":353,"children":354},{},[355,357,363],{"type":48,"value":356},"All LangGraph graphs use ",{"type":43,"tag":176,"props":358,"children":360},{"className":359},[],[361],{"type":48,"value":362},"StateGraph(State)",{"type":48,"value":364}," with explicit nodes, edges, and conditional edges.",{"type":43,"tag":366,"props":367,"children":368},"deep-agents-profile",{},[369,375,382,410,417,430],{"type":43,"tag":221,"props":370,"children":372},{"id":371},"deep-agents-agent-harness",[373],{"type":48,"value":374},"Deep Agents — agent harness",{"type":43,"tag":117,"props":376,"children":377},{},[378],{"type":43,"tag":59,"props":379,"children":380},{},[381],{"type":48,"value":234},{"type":43,"tag":51,"props":383,"children":384},{},[385,390,395,400,405],{"type":43,"tag":55,"props":386,"children":387},{},[388],{"type":48,"value":389},"Long-running tasks that require planning and decomposition",{"type":43,"tag":55,"props":391,"children":392},{},[393],{"type":48,"value":394},"Agents that read, write, and manage files across a session",{"type":43,"tag":55,"props":396,"children":397},{},[398],{"type":48,"value":399},"Delegating subtasks to specialized subagents",{"type":43,"tag":55,"props":401,"children":402},{},[403],{"type":48,"value":404},"Persistent memory across sessions",{"type":43,"tag":55,"props":406,"children":407},{},[408],{"type":48,"value":409},"Loading domain-specific skills on demand",{"type":43,"tag":117,"props":411,"children":412},{},[413],{"type":43,"tag":59,"props":414,"children":415},{},[416],{"type":48,"value":260},{"type":43,"tag":51,"props":418,"children":419},{},[420,425],{"type":43,"tag":55,"props":421,"children":422},{},[423],{"type":48,"value":424},"The task is simple enough for a single-purpose agent",{"type":43,"tag":55,"props":426,"children":427},{},[428],{"type":48,"value":429},"You need precise hand-crafted control over every graph edge (use LangGraph directly)",{"type":43,"tag":117,"props":431,"children":432},{},[433,435,441],{"type":48,"value":434},"All Deep Agents use ",{"type":43,"tag":176,"props":436,"children":438},{"className":437},[],[439],{"type":48,"value":440},"create_deep_agent(model, tools=[...])",{"type":48,"value":291},{"type":43,"tag":123,"props":443,"children":444},{},[],{"type":43,"tag":127,"props":446,"children":448},{"id":447},"mixing-layers",[449],{"type":48,"value":450},"Mixing Layers",{"type":43,"tag":447,"props":452,"children":453},{},[454,459,482],{"type":43,"tag":117,"props":455,"children":456},{},[457],{"type":48,"value":458},"The tools are layered, so they can be combined in the same project. Common patterns:",{"type":43,"tag":51,"props":460,"children":461},{},[462,472],{"type":43,"tag":55,"props":463,"children":464},{},[465,470],{"type":43,"tag":59,"props":466,"children":467},{},[468],{"type":48,"value":469},"Deep Agents orchestrator → LangGraph subagent",{"type":48,"value":471}," — when the main agent needs planning and memory but one subtask requires a deterministic graph.",{"type":43,"tag":55,"props":473,"children":474},{},[475,480],{"type":43,"tag":59,"props":476,"children":477},{},[478],{"type":48,"value":479},"LangGraph graph wrapped as a tool or subagent",{"type":48,"value":481}," — when a specialized pipeline (e.g. RAG, reflection loop) is called by a broader agent.",{"type":43,"tag":117,"props":483,"children":484},{},[485,487,493],{"type":48,"value":486},"A compiled LangGraph graph can be registered as a named subagent inside Deep Agents — the orchestrator delegates to it via the ",{"type":43,"tag":176,"props":488,"children":490},{"className":489},[],[491],{"type":48,"value":492},"task",{"type":48,"value":494}," tool without knowing its internal structure. LangChain tools and retrievers work freely inside both LangGraph nodes and Deep Agents tools.",{"type":43,"tag":123,"props":496,"children":497},{},[],{"type":43,"tag":127,"props":499,"children":501},{"id":500},"step-2-set-environment-variables",[502],{"type":48,"value":503},"Step 2 — Set Environment Variables",{"type":43,"tag":117,"props":505,"children":506},{},[507],{"type":48,"value":508},"Always set these for observability. These are the current LangSmith env var names. Copy them as-is. OLDER NAMES NO LONGER WORK.",{"type":43,"tag":510,"props":511,"children":512},"environment-variables",{},[513,515],{"type":48,"value":514},"\nLANGSMITH_API_KEY=",{"type":43,"tag":516,"props":517,"children":518},"your-key",{},[519,521],{"type":48,"value":520},"\nLANGSMITH_TRACING=true\nLANGSMITH_PROJECT=",{"type":43,"tag":522,"props":523,"children":524},"project-name",{},[],{"type":43,"tag":117,"props":526,"children":527},{},[528,530,536,538,544,545,551],{"type":48,"value":529},"Model-provider and tool-specific keys (",{"type":43,"tag":176,"props":531,"children":533},{"className":532},[],[534],{"type":48,"value":535},"ANTHROPIC_API_KEY",{"type":48,"value":537},", ",{"type":43,"tag":176,"props":539,"children":541},{"className":540},[],[542],{"type":48,"value":543},"OPENAI_API_KEY",{"type":48,"value":537},{"type":43,"tag":176,"props":546,"children":548},{"className":547},[],[549],{"type":48,"value":550},"TAVILY_API_KEY",{"type":48,"value":552},", etc.) depend on your stack — set them as needed.",{"type":43,"tag":123,"props":554,"children":555},{},[],{"type":43,"tag":127,"props":557,"children":559},{"id":558},"step-3-how-the-docs-work",[560],{"type":48,"value":561},"Step 3 — How the Docs Work",{"type":43,"tag":563,"props":564,"children":565},"docs",{},[566,578,616,621,627,648,710,716,733,745,763,789],{"type":43,"tag":117,"props":567,"children":568},{},[569,571,576],{"type":48,"value":570},"All documentation lives at ",{"type":43,"tag":59,"props":572,"children":573},{},[574],{"type":48,"value":575},"docs.langchain.com",{"type":48,"value":577},", organized into two top-level sections:",{"type":43,"tag":51,"props":579,"children":580},{},[581,607],{"type":43,"tag":55,"props":582,"children":583},{},[584,589,591,597,599,605],{"type":43,"tag":59,"props":585,"children":586},{},[587],{"type":48,"value":588},"OSS",{"type":48,"value":590}," — LangChain, LangGraph, Deep Agents. Python (",{"type":43,"tag":176,"props":592,"children":594},{"className":593},[],[595],{"type":48,"value":596},"\u002Foss\u002Fpython\u002F",{"type":48,"value":598},") and TypeScript (",{"type":43,"tag":176,"props":600,"children":602},{"className":601},[],[603],{"type":48,"value":604},"\u002Foss\u002Fjavascript\u002F",{"type":48,"value":606},") trees in parallel.",{"type":43,"tag":55,"props":608,"children":609},{},[610,614],{"type":43,"tag":59,"props":611,"children":612},{},[613],{"type":48,"value":113},{"type":48,"value":615}," — observability, evaluation, deployment, prompt engineering.",{"type":43,"tag":117,"props":617,"children":618},{},[619],{"type":48,"value":620},"Each product has its own page tree: overview → quickstart → how-to guides → reference.",{"type":43,"tag":221,"props":622,"children":624},{"id":623},"canonical-landing-pages",[625],{"type":48,"value":626},"Canonical landing pages",{"type":43,"tag":117,"props":628,"children":629},{},[630,632,638,640,646],{"type":48,"value":631},"Start here rather than tree-searching from root (swap ",{"type":43,"tag":176,"props":633,"children":635},{"className":634},[],[636],{"type":48,"value":637},"python",{"type":48,"value":639}," → ",{"type":43,"tag":176,"props":641,"children":643},{"className":642},[],[644],{"type":48,"value":645},"javascript",{"type":48,"value":647}," for TypeScript):",{"type":43,"tag":51,"props":649,"children":650},{},[651,666,680,694],{"type":43,"tag":55,"props":652,"children":653},{},[654,658,660],{"type":43,"tag":59,"props":655,"children":656},{},[657],{"type":48,"value":9},{"type":48,"value":659}," — ",{"type":43,"tag":176,"props":661,"children":663},{"className":662},[],[664],{"type":48,"value":665},"\u002Foss\u002Fpython\u002Flangchain\u002Foverview",{"type":43,"tag":55,"props":667,"children":668},{},[669,673,674],{"type":43,"tag":59,"props":670,"children":671},{},[672],{"type":48,"value":19},{"type":48,"value":659},{"type":43,"tag":176,"props":675,"children":677},{"className":676},[],[678],{"type":48,"value":679},"\u002Foss\u002Fpython\u002Flanggraph\u002Foverview",{"type":43,"tag":55,"props":681,"children":682},{},[683,687,688],{"type":43,"tag":59,"props":684,"children":685},{},[686],{"type":48,"value":63},{"type":48,"value":659},{"type":43,"tag":176,"props":689,"children":691},{"className":690},[],[692],{"type":48,"value":693},"\u002Foss\u002Fpython\u002Fdeepagents\u002Foverview",{"type":43,"tag":55,"props":695,"children":696},{},[697,701,702,708],{"type":43,"tag":59,"props":698,"children":699},{},[700],{"type":48,"value":113},{"type":48,"value":659},{"type":43,"tag":176,"props":703,"children":705},{"className":704},[],[706],{"type":48,"value":707},"\u002Flangsmith\u002Fhome",{"type":48,"value":709}," (no language split)",{"type":43,"tag":221,"props":711,"children":713},{"id":712},"accessing-docs-in-an-agent-context",[714],{"type":48,"value":715},"Accessing docs in an agent context",{"type":43,"tag":117,"props":717,"children":718},{},[719,724,725,731],{"type":43,"tag":59,"props":720,"children":721},{},[722],{"type":48,"value":723},"If the LangChain Docs MCP server is connected",{"type":48,"value":174},{"type":43,"tag":176,"props":726,"children":728},{"className":727},[],[729],{"type":48,"value":730},"mcp__docs-langchain__*",{"type":48,"value":732}," tools are available), query it directly:",{"type":43,"tag":734,"props":735,"children":739},"pre",{"className":736,"code":738,"language":48},[737],"language-text","tree \u002Foss\u002Fpython -L 2                        # explore Python structure\ntree \u002Foss\u002Fjavascript -L 2                    # parallel TypeScript structure\ncat \u002Foss\u002Fpython\u002Flangchain\u002Fquickstart.mdx     # read a specific page\nrg -il \"checkpointer\" \u002Foss\u002Fpython\u002Flanggraph\u002F # search by keyword\n",[740],{"type":43,"tag":176,"props":741,"children":743},{"__ignoreMap":742},"",[744],{"type":48,"value":738},{"type":43,"tag":117,"props":746,"children":747},{},[748,753,755,761],{"type":43,"tag":59,"props":749,"children":750},{},[751],{"type":48,"value":752},"If the MCP server is not available",{"type":48,"value":754},", use the ",{"type":43,"tag":176,"props":756,"children":758},{"className":757},[],[759],{"type":48,"value":760},"llms.txt",{"type":48,"value":762}," index:",{"type":43,"tag":143,"props":764,"children":765},{},[766,779,784],{"type":43,"tag":55,"props":767,"children":768},{},[769,771,777],{"type":48,"value":770},"Fetch ",{"type":43,"tag":176,"props":772,"children":774},{"className":773},[],[775],{"type":48,"value":776},"https:\u002F\u002Fdocs.langchain.com\u002Fllms.txt",{"type":48,"value":778}," — structured list of all pages with descriptions",{"type":43,"tag":55,"props":780,"children":781},{},[782],{"type":48,"value":783},"Identify the 2–4 most relevant pages for the question",{"type":43,"tag":55,"props":785,"children":786},{},[787],{"type":48,"value":788},"Fetch those pages directly for accurate, up-to-date content",{"type":43,"tag":790,"props":791,"children":792},"blockquote",{},[793],{"type":43,"tag":117,"props":794,"children":795},{},[796],{"type":48,"value":797},"Always prefer fetching live docs over relying on training-data knowledge — these libraries evolve fast and APIs change often.",{"type":43,"tag":123,"props":799,"children":800},{},[],{"type":43,"tag":127,"props":802,"children":804},{"id":803},"step-4-load-the-right-skill-next",[805],{"type":48,"value":806},"Step 4 — Load the Right Skill Next",{"type":43,"tag":117,"props":808,"children":809},{},[810],{"type":48,"value":811},"If the user only wants a minimal local working agent (new project, stub tool, provider key), load the matching quickstart first:",{"type":43,"tag":51,"props":813,"children":814},{},[815,834,852],{"type":43,"tag":55,"props":816,"children":817},{},[818,820,826,828],{"type":48,"value":819},"LangChain → ",{"type":43,"tag":176,"props":821,"children":823},{"className":822},[],[824],{"type":48,"value":825},"langchain-python-quickstart",{"type":48,"value":827}," or ",{"type":43,"tag":176,"props":829,"children":831},{"className":830},[],[832],{"type":48,"value":833},"langchain-typescript-quickstart",{"type":43,"tag":55,"props":835,"children":836},{},[837,839,845,846],{"type":48,"value":838},"LangGraph → ",{"type":43,"tag":176,"props":840,"children":842},{"className":841},[],[843],{"type":48,"value":844},"langgraph-python-quickstart",{"type":48,"value":827},{"type":43,"tag":176,"props":847,"children":849},{"className":848},[],[850],{"type":48,"value":851},"langgraph-typescript-quickstart",{"type":43,"tag":55,"props":853,"children":854},{},[855,857,863,864],{"type":48,"value":856},"Deep Agents → ",{"type":43,"tag":176,"props":858,"children":860},{"className":859},[],[861],{"type":48,"value":862},"deepagents-python-quickstart",{"type":48,"value":827},{"type":43,"tag":176,"props":865,"children":867},{"className":866},[],[868],{"type":48,"value":869},"deepagents-typescript-quickstart",{"type":43,"tag":117,"props":871,"children":872},{},[873],{"type":48,"value":874},"Otherwise load the skill below that matches your layer from Step 1. This is required — the layer-specific skill carries the current API; the primer alone does not.",{"type":43,"tag":876,"props":877,"children":878},"next-skills",{},[879,883,942,946,991,996,1014],{"type":43,"tag":221,"props":880,"children":881},{"id":8},[882],{"type":48,"value":9},{"type":43,"tag":51,"props":884,"children":885},{},[886,900,914,928],{"type":43,"tag":55,"props":887,"children":888},{},[889,898],{"type":43,"tag":59,"props":890,"children":891},{},[892],{"type":43,"tag":176,"props":893,"children":895},{"className":894},[],[896],{"type":48,"value":897},"langchain-fundamentals",{"type":48,"value":899}," — building any LangChain agent",{"type":43,"tag":55,"props":901,"children":902},{},[903,912],{"type":43,"tag":59,"props":904,"children":905},{},[906],{"type":43,"tag":176,"props":907,"children":909},{"className":908},[],[910],{"type":48,"value":911},"langchain-rag",{"type":48,"value":913}," — adding RAG \u002F vector store retrieval",{"type":43,"tag":55,"props":915,"children":916},{},[917,926],{"type":43,"tag":59,"props":918,"children":919},{},[920],{"type":43,"tag":176,"props":921,"children":923},{"className":922},[],[924],{"type":48,"value":925},"langchain-middleware",{"type":48,"value":927}," — structured output with Pydantic",{"type":43,"tag":55,"props":929,"children":930},{},[931,940],{"type":43,"tag":59,"props":932,"children":933},{},[934],{"type":43,"tag":176,"props":935,"children":937},{"className":936},[],[938],{"type":48,"value":939},"langchain-dependencies",{"type":48,"value":941}," — package versions, installs, or dependency management questions",{"type":43,"tag":221,"props":943,"children":944},{"id":20},[945],{"type":48,"value":19},{"type":43,"tag":51,"props":947,"children":948},{},[949,963,977],{"type":43,"tag":55,"props":950,"children":951},{},[952,961],{"type":43,"tag":59,"props":953,"children":954},{},[955],{"type":43,"tag":176,"props":956,"children":958},{"className":957},[],[959],{"type":48,"value":960},"langgraph-fundamentals",{"type":48,"value":962}," — any LangGraph graph",{"type":43,"tag":55,"props":964,"children":965},{},[966,975],{"type":43,"tag":59,"props":967,"children":968},{},[969],{"type":43,"tag":176,"props":970,"children":972},{"className":971},[],[973],{"type":48,"value":974},"langgraph-human-in-the-loop",{"type":48,"value":976}," — human-in-the-loop or approval workflows",{"type":43,"tag":55,"props":978,"children":979},{},[980,989],{"type":43,"tag":59,"props":981,"children":982},{},[983],{"type":43,"tag":176,"props":984,"children":986},{"className":985},[],[987],{"type":48,"value":988},"langgraph-persistence",{"type":48,"value":990}," — state that must survive restarts, or cross-thread memory",{"type":43,"tag":221,"props":992,"children":994},{"id":993},"deep-agents",[995],{"type":48,"value":63},{"type":43,"tag":117,"props":997,"children":998},{},[999,1012],{"type":43,"tag":59,"props":1000,"children":1001},{},[1002,1004,1010],{"type":48,"value":1003},"Always load ",{"type":43,"tag":176,"props":1005,"children":1007},{"className":1006},[],[1008],{"type":48,"value":1009},"deep-agents-core",{"type":48,"value":1011}," first.",{"type":48,"value":1013}," Then, as needed:",{"type":43,"tag":51,"props":1015,"children":1016},{},[1017,1031],{"type":43,"tag":55,"props":1018,"children":1019},{},[1020,1029],{"type":43,"tag":59,"props":1021,"children":1022},{},[1023],{"type":43,"tag":176,"props":1024,"children":1026},{"className":1025},[],[1027],{"type":48,"value":1028},"deep-agents-orchestration",{"type":48,"value":1030}," — subagent delegation or orchestration",{"type":43,"tag":55,"props":1032,"children":1033},{},[1034,1043],{"type":43,"tag":59,"props":1035,"children":1036},{},[1037],{"type":43,"tag":176,"props":1038,"children":1040},{"className":1039},[],[1041],{"type":48,"value":1042},"deep-agents-memory",{"type":48,"value":1044}," — cross-session persistent memory",{"items":1046,"total":1221},[1047,1068,1079,1096,1109,1124,1139,1154,1168,1178,1189,1208],{"slug":1048,"name":1048,"fn":1049,"description":1050,"org":1051,"tags":1052,"stars":1065,"repoUrl":1066,"updatedAt":1067},"analyze-market","perform market analysis and size estimation","Perform a market analysis for a product category or segment. Trigger on: market analysis, market size, TAM SAM SOM, market opportunity, industry analysis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1053,1056,1059,1062],{"name":1054,"slug":1055,"type":16},"Marketing","marketing",{"name":1057,"slug":1058,"type":16},"Research","research",{"name":1060,"slug":1061,"type":16},"Sales","sales",{"name":1063,"slug":1064,"type":16},"Strategy","strategy",26592,"https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Fdeepagents","2026-04-18T04:46:54.557115",{"slug":1069,"name":1069,"fn":1070,"description":1071,"org":1072,"tags":1073,"stars":1065,"repoUrl":1066,"updatedAt":1078},"arxiv-search","search arXiv for academic research papers","Searches arXiv for preprints and academic papers, retrieves abstracts, and filters by topic. Use when the user asks to find research papers, search arXiv, look up preprints, find academic articles in physics, math, CS, biology, statistics, or related fields.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1074,1075],{"name":1057,"slug":1058,"type":16},{"name":1076,"slug":1077,"type":16},"Search","search","2026-05-13T06:11:01.203061",{"slug":1080,"name":1080,"fn":1081,"description":1082,"org":1083,"tags":1084,"stars":1065,"repoUrl":1066,"updatedAt":1095},"blog-post","write SEO-optimized blog posts","Write long-form blog posts with SEO optimization and clear structure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1085,1088,1089,1092],{"name":1086,"slug":1087,"type":16},"Content Creation","content-creation",{"name":1054,"slug":1055,"type":16},{"name":1090,"slug":1091,"type":16},"SEO","seo",{"name":1093,"slug":1094,"type":16},"Writing","writing","2026-04-15T05:00:54.149813",{"slug":1097,"name":1097,"fn":1098,"description":1099,"org":1100,"tags":1101,"stars":1065,"repoUrl":1066,"updatedAt":1108},"competitor-analysis","analyze competitors and market positioning","Analyze competitors in a given market segment. Trigger on: competitive landscape, competitor analysis, market comparison, competitive positioning.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1102,1105,1106,1107],{"name":1103,"slug":1104,"type":16},"Competitive Intelligence","competitive-intelligence",{"name":1054,"slug":1055,"type":16},{"name":1057,"slug":1058,"type":16},{"name":1063,"slug":1064,"type":16},"2026-04-18T04:46:55.79306",{"slug":1110,"name":1110,"fn":1111,"description":1112,"org":1113,"tags":1114,"stars":1065,"repoUrl":1066,"updatedAt":1123},"deepagents-thread-inspector","inspect local Deep Agents conversation threads","Inspect and explain conversations in the local Deep Agents Code SQLite session store. Use as a fallback when LangSmith trace tooling is unavailable, for offline or untraced sessions, or when asked to identify or summarize a local dcode thread, inspect checkpoint metadata, list recent local threads, or parse ~\u002F.deepagents\u002F.state\u002Fsessions.db and a thread UUID or prefix.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1115,1116,1119,1120],{"name":22,"slug":23,"type":16},{"name":1117,"slug":1118,"type":16},"Debugging","debugging",{"name":9,"slug":8,"type":16},{"name":1121,"slug":1122,"type":16},"SQLite","sqlite","2026-07-24T06:08:57.102689",{"slug":1125,"name":1125,"fn":1126,"description":1127,"org":1128,"tags":1129,"stars":1065,"repoUrl":1066,"updatedAt":1138},"langgraph-docs","build stateful agents with LangGraph","Fetches and references LangGraph Python documentation to build stateful agents, create multi-agent workflows, and implement human-in-the-loop patterns. Use when the user asks about LangGraph, graph agents, state machines, agent orchestration, LangGraph API, or needs LangGraph implementation guidance.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1130,1131,1134,1135],{"name":22,"slug":23,"type":16},{"name":1132,"slug":1133,"type":16},"Documentation","documentation",{"name":19,"slug":20,"type":16},{"name":1136,"slug":1137,"type":16},"Multi-Agent","multi-agent","2026-05-13T06:11:03.650877",{"slug":1140,"name":1140,"fn":1141,"description":1142,"org":1143,"tags":1144,"stars":1065,"repoUrl":1066,"updatedAt":1153},"remember","capture knowledge into persistent memory","Review the current conversation and capture valuable knowledge — best practices, coding conventions, architecture decisions, workflows, and user feedback — into persistent memory (AGENTS.md) or reusable skills. Use when the user says: (1) remember this, (2) save what we learned, (3) update memory, (4) capture learnings.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1145,1146,1147,1150],{"name":22,"slug":23,"type":16},{"name":1132,"slug":1133,"type":16},{"name":1148,"slug":1149,"type":16},"Knowledge Management","knowledge-management",{"name":1151,"slug":1152,"type":16},"Memory","memory","2026-05-13T06:10:58.510037",{"slug":1155,"name":1155,"fn":1156,"description":1157,"org":1158,"tags":1159,"stars":1065,"repoUrl":1066,"updatedAt":1167},"skill-creator","create agent skills and tool integrations","Guide for creating effective skills that extend agent capabilities with specialized knowledge, workflows, or tool integrations. Use this skill when the user asks to: (1) create a new skill, (2) make a skill, (3) build a skill, (4) set up a skill, (5) initialize a skill, (6) scaffold a skill, (7) update or modify an existing skill, (8) validate a skill, (9) learn about skill structure, (10) understand how skills work, or (11) get guidance on skill design patterns. Trigger on phrases like \"create a skill\", \"new skill\", \"make a skill\", \"skill for X\", \"how do I create a skill\", or \"help me build a skill\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1160,1161,1164],{"name":22,"slug":23,"type":16},{"name":1162,"slug":1163,"type":16},"Engineering","engineering",{"name":1165,"slug":1166,"type":16},"Plugin Development","plugin-development","2026-05-13T06:10:59.88449",{"slug":1169,"name":1169,"fn":1170,"description":1171,"org":1172,"tags":1173,"stars":1065,"repoUrl":1066,"updatedAt":1177},"social-media","create optimized social media posts","Create social media posts optimized for engagement across platforms.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1174,1175,1176],{"name":1086,"slug":1087,"type":16},{"name":1054,"slug":1055,"type":16},{"name":1093,"slug":1094,"type":16},"2026-04-15T05:00:55.37452",{"slug":1179,"name":1179,"fn":1180,"description":1181,"org":1182,"tags":1183,"stars":1065,"repoUrl":1066,"updatedAt":1188},"web-research","conduct and synthesize web research","Searches multiple web sources, synthesizes findings, and produces cited research reports using delegated subagents. Use when the user asks to research a topic online, search the web, look something up, find current information, compare options, or produce a research report.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1184,1185,1186,1187],{"name":22,"slug":23,"type":16},{"name":1136,"slug":1137,"type":16},{"name":1057,"slug":1058,"type":16},{"name":1076,"slug":1077,"type":16},"2026-05-13T06:11:04.930044",{"slug":1190,"name":1190,"fn":1191,"description":1192,"org":1193,"tags":1194,"stars":1205,"repoUrl":1206,"updatedAt":1207},"mermaid-diagrams","embed Mermaid diagrams in documentation","Embed Mermaid diagrams in generated wiki pages. Use whenever documenting a runtime or request flow, a call sequence, a state machine or lifecycle, a data model or entity relationships, or non-trivial control flow, since these are clearer as a diagram than as prose. Also use when an update run touches a page that already contains a mermaid fence, or a page that contains a text fence a previous run degraded.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1195,1198,1199,1202],{"name":1196,"slug":1197,"type":16},"Diagrams","diagrams",{"name":1132,"slug":1133,"type":16},{"name":1200,"slug":1201,"type":16},"Markdown","markdown",{"name":1203,"slug":1204,"type":16},"Technical Writing","technical-writing",12181,"https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Fopenwiki","2026-07-24T06:09:01.089597",{"slug":1209,"name":1209,"fn":1210,"description":1211,"org":1212,"tags":1213,"stars":1205,"repoUrl":1206,"updatedAt":1220},"write-connector","implement OpenWiki source connectors","Add a new built-in OpenWiki source connector. Use when a user asks to create or implement an OpenWiki connector.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1214,1217],{"name":1215,"slug":1216,"type":16},"API Development","api-development",{"name":1218,"slug":1219,"type":16},"Integrations","integrations","2026-07-18T05:48:23.961804",41,{"items":1223,"total":1310},[1224,1236,1248,1263,1273,1284,1292],{"slug":1009,"name":1009,"fn":1225,"description":1226,"org":1227,"tags":1228,"stars":27,"repoUrl":28,"updatedAt":1235},"build applications with LangChain Deep Agents","INVOKE THIS SKILL when building ANY Deep Agents application. Covers create_deep_agent(), harness architecture, SKILL.md format, and configuration options.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1229,1230,1231,1232],{"name":22,"slug":23,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"name":1233,"slug":1234,"type":16},"LLM","llm","2026-04-06T18:26:24.822592",{"slug":1042,"name":1042,"fn":1237,"description":1238,"org":1239,"tags":1240,"stars":27,"repoUrl":28,"updatedAt":1247},"implement memory and persistence for Deep Agents","INVOKE THIS SKILL when your Deep Agent needs memory, persistence, or filesystem access. Covers StateBackend (ephemeral), StoreBackend (persistent), FilesystemMiddleware, and CompositeBackend for routing.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1241,1242,1243,1244],{"name":22,"slug":23,"type":16},{"name":9,"slug":8,"type":16},{"name":1151,"slug":1152,"type":16},{"name":1245,"slug":1246,"type":16},"Storage","storage","2026-04-06T18:26:26.065654",{"slug":1028,"name":1028,"fn":1249,"description":1250,"org":1251,"tags":1252,"stars":27,"repoUrl":28,"updatedAt":1262},"orchestrate subagents and tasks in Deep Agents","INVOKE THIS SKILL when using subagents, task planning, or human approval in Deep Agents. Covers SubAgentMiddleware, TodoList for planning, and HITL interrupts.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1253,1254,1257,1258,1259],{"name":22,"slug":23,"type":16},{"name":1255,"slug":1256,"type":16},"Approvals","approvals",{"name":9,"slug":8,"type":16},{"name":1136,"slug":1137,"type":16},{"name":1260,"slug":1261,"type":16},"Workflow Automation","workflow-automation","2026-04-06T18:26:32.280463",{"slug":862,"name":862,"fn":1264,"description":1265,"org":1266,"tags":1267,"stars":27,"repoUrl":28,"updatedAt":1272},"scaffold local Deep Agents","Scaffold a minimal local Deep Agent in Python by following the official quickstart, using provider-native web search instead of Tavily. Use when the user wants to quickly build or try a Deep Agent locally.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1268,1269,1270],{"name":22,"slug":23,"type":16},{"name":9,"slug":8,"type":16},{"name":1271,"slug":637,"type":16},"Python","2026-07-24T06:09:00.291108",{"slug":869,"name":869,"fn":1274,"description":1275,"org":1276,"tags":1277,"stars":27,"repoUrl":28,"updatedAt":1283},"build Deep Agents in TypeScript","Scaffold a minimal local Deep Agent in TypeScript by following the official quickstart, using provider-native web search instead of Tavily. Use when the user wants to quickly build or try a Deep Agent locally.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1278,1279,1280],{"name":22,"slug":23,"type":16},{"name":9,"slug":8,"type":16},{"name":1281,"slug":1282,"type":16},"TypeScript","typescript","2026-07-24T06:09:00.673714",{"slug":4,"name":4,"fn":5,"description":6,"org":1285,"tags":1286,"stars":27,"repoUrl":28,"updatedAt":29},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1287,1288,1289,1290,1291],{"name":22,"slug":23,"type":16},{"name":25,"slug":26,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"name":19,"slug":20,"type":16},{"slug":1293,"name":1293,"fn":1294,"description":1295,"org":1296,"tags":1297,"stars":27,"repoUrl":28,"updatedAt":1309},"eval-engineering","create and audit Harbor agent evals","Iteratively inspect an agent repository and optional user-provided traces, interview the user, and create, run, and audit Harbor evals one at a time. Use for agent evals, Harbor tasks, benchmark cases, verifier design, or controlled agent environments.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1298,1299,1302,1305,1306],{"name":22,"slug":23,"type":16},{"name":1300,"slug":1301,"type":16},"Code Analysis","code-analysis",{"name":1303,"slug":1304,"type":16},"Evals","evals",{"name":9,"slug":8,"type":16},{"name":1307,"slug":1308,"type":16},"Testing","testing","2026-07-31T05:53:29.552458",22]