langchain-rag

Lets an assistant answer from your own documents: they are split into passages, indexed by meaning rather than by keyword, and the passages that fit the question are pulled back in as it answers.

WHEN TO USE

SCOPE

USE FOR

Build a pipeline that loads, chunks, embeds, and retrieves documents to answer a query.
Load PDFs, web pages, or a directory of text files as documents to index.
Filter search results by metadata, such as returning only documents in one language.
Search in a way that returns relevant results without near-duplicate repeats.
Give an agent a tool that searches your own documents instead of the open web.

NOT FOR

Never load a downloaded or otherwise untrusted FAISS index with dangerous deserialization enabled.
Never use different embedding models for indexing and querying in the same vector store.

This is a saved copy of the skill's SKILL.md instructions. It may differ from the upstream file or the version you install.

In this document (5)

Retrieval Augmented Generation (RAG) enhances LLM responses by fetching relevant context from external knowledge sources.

Pipeline:

Index: Load → Split → Embed → Store
Retrieve: Query → Embed → Search → Return docs
Generate: Docs + Query → LLM → Response

Key Components:

Document Loaders: Ingest data from files, web, databases
Text Splitters: Break documents into chunks
Embeddings: Convert text to vectors
Vector Stores: Store and search embeddings

Vector Store	Use Case	Persistence
InMemory	Testing	Memory only
FAISS	Local, high performance	Disk
Chroma	Development	Disk
Pinecone	Production, managed	Cloud

Complete RAG Pipeline

End-to-end RAG pipeline: load documents, split into chunks, embed, store, retrieve, and generate a response.

from langchain_openai import ChatOpenAI, OpenAIEmbeddings
from langchain_community.vectorstores import InMemoryVectorStore
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_core.documents import Document

# 1. Load documents
docs = [
    Document(page_content="LangChain is a framework for LLM apps.", metadata={}),
    Document(page_content="RAG = Retrieval Augmented Generation.", metadata={}),
]

# 2. Split documents
splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
splits = splitter.split_documents(docs)

# 3. Create embeddings and store
embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
vectorstore = InMemoryVectorStore.from_documents(splits, embeddings)

# 4. Create retriever
retriever = vectorstore.as_retriever(search_kwargs={"k": 4})

# 5. Use in RAG
model = ChatOpenAI(model="gpt-4.1")
query = "What is RAG?"
relevant_docs = retriever.invoke(query)

context = "\n\n".join([doc.page_content for doc in relevant_docs])
response = model.invoke([
    {"role": "system", "content": f"Use this context:\n\n{context}"},
    {"role": "user", "content": query},
])

End-to-end RAG pipeline: load documents, split into chunks, embed, store, retrieve, and generate a response.

import { ChatOpenAI, OpenAIEmbeddings } from "@langchain/openai";
import { MemoryVectorStore } from "@langchain/classic/vectorstores/memory";
import { RecursiveCharacterTextSplitter } from "@langchain/textsplitters";
import { Document } from "@langchain/core/documents";

// 1. Load documents
const docs = [
  new Document({ pageContent: "LangChain is a framework for LLM apps.", metadata: {} }),
  new Document({ pageContent: "RAG = Retrieval Augmented Generation.", metadata: {} }),
];

// 2. Split documents
const splitter = new RecursiveCharacterTextSplitter({ chunkSize: 500, chunkOverlap: 50 });
const splits = await splitter.splitDocuments(docs);

// 3. Create embeddings and store
const embeddings = new OpenAIEmbeddings({ model: "text-embedding-3-small" });
const vectorstore = await MemoryVectorStore.fromDocuments(splits, embeddings);

// 4. Create retriever
const retriever = vectorstore.asRetriever({ k: 4 });

// 5. Use in RAG
const model = new ChatOpenAI({ model: "gpt-4.1" });
const query = "What is RAG?";
const relevantDocs = await retriever.invoke(query);

const context = relevantDocs.map(doc => doc.pageContent).join("\n\n");
const response = await model.invoke([
  { role: "system", content: `Use this context:\n\n${context}` },
  { role: "user", content: query },
]);

Document Loaders

Load a PDF file and extract each page as a separate document.

from langchain_community.document_loaders import PyPDFLoader

loader = PyPDFLoader("./document.pdf")
docs = loader.load()
print(f"Loaded {len(docs)} pages")

Load a PDF file and extract each page as a separate document.

import { PDFLoader } from "@langchain/community/document_loaders/fs/pdf";

const loader = new PDFLoader("./document.pdf");
const docs = await loader.load();
console.log(`Loaded ${docs.length} pages`);

Fetch and parse content from a web URL into a document.

from langchain_community.document_loaders import WebBaseLoader

loader = WebBaseLoader("https://docs.langchain.com")
docs = loader.load()

Fetch and parse content from a web URL into a document using Cheerio.

import { CheerioWebBaseLoader } from "@langchain/community/document_loaders/web/cheerio";

const loader = new CheerioWebBaseLoader("https://docs.langchain.com");
const docs = await loader.load();

Load all text files from a directory using a glob pattern.

from langchain_community.document_loaders import DirectoryLoader, TextLoader

# Load all text files from directory
loader = DirectoryLoader(
    "path/to/documents",
    glob="**/*.txt",  # Pattern for files to load
    loader_cls=TextLoader
)
docs = loader.load()

Text Splitting

Split documents into chunks using RecursiveCharacterTextSplitter with configurable size and overlap.

from langchain_text_splitters import RecursiveCharacterTextSplitter

splitter = RecursiveCharacterTextSplitter(
    chunk_size=1000,        # Characters per chunk
    chunk_overlap=200,      # Overlap for context continuity
    separators=["\n\n", "\n", " ", ""],  # Split hierarchy
)

splits = splitter.split_documents(docs)

Vector Stores

Create a persistent Chroma vector store and reload it from disk.

from langchain_chroma import Chroma
from langchain_openai import OpenAIEmbeddings

vectorstore = Chroma.from_documents(
    documents=splits,
    embedding=OpenAIEmbeddings(),
    persist_directory="./chroma_db",
    collection_name="my-collection",
)

# Load existing
vectorstore = Chroma(
    persist_directory="./chroma_db",
    embedding_function=OpenAIEmbeddings(),
    collection_name="my-collection",
)

Create a Chroma vector store connected to a running Chroma server.

import { Chroma } from "@langchain/community/vectorstores/chroma";
import { OpenAIEmbeddings } from "@langchain/openai";

const vectorstore = await Chroma.fromDocuments(
  splits,
  new OpenAIEmbeddings(),
  { collectionName: "my-collection", url: "http://localhost:8000" }
);

Create a FAISS vector store, save it to disk, and reload it.

from langchain_community.vectorstores import FAISS

vectorstore = FAISS.from_documents(splits, embeddings)
vectorstore.save_local("./faiss_index")

# Only load FAISS indexes that you created and fully control.
# The Python FAISS loader uses pickle-backed metadata, so never load
# downloaded, shared, or otherwise untrusted index directories.
loaded = FAISS.load_local(
    "./faiss_index",
    embeddings,
    allow_dangerous_deserialization=True,
)

Create a FAISS vector store, save it to disk, and reload it.

import { FaissStore } from "@langchain/community/vectorstores/faiss";

const vectorstore = await FaissStore.fromDocuments(splits, embeddings);
await vectorstore.save("./faiss_index");

const loaded = await FaissStore.load("./faiss_index", embeddings);

Retrieval

Perform similarity search and retrieve results with relevance scores.

# Basic search
results = vectorstore.similarity_search(query, k=5)

# With scores
results_with_score = vectorstore.similarity_search_with_score(query, k=5)
for doc, score in results_with_score:
    print(f"Score: {score}, Content: {doc.page_content}")

Perform similarity search and retrieve results with relevance scores.

// Basic search
const results = await vectorstore.similaritySearch(query, 5);

// With scores
const resultsWithScore = await vectorstore.similaritySearchWithScore(query, 5);
for (const [doc, score] of resultsWithScore) {
  console.log(`Score: ${score}, Content: ${doc.pageContent}`);
}

Use MMR (Maximal Marginal Relevance) to balance relevance and diversity in search results.

# MMR balances relevance and diversity
retriever = vectorstore.as_retriever(
    search_type="mmr",
    search_kwargs={"fetch_k": 20, "lambda_mult": 0.5, "k": 5},
)

Add metadata to documents and filter search results by metadata properties.

# Add metadata when creating documents
docs = [
    Document(
        page_content="Python programming guide",
        metadata={"language": "python", "topic": "programming"}
    ),
]

# Search with filter
results = vectorstore.similarity_search(
    "programming",
    k=5,
    filter={"language": "python"}  # Only Python docs
)

Create an agent that uses RAG as a tool for answering questions.

from langchain.agents import create_agent
from langchain.tools import tool

@tool
def search_docs(query: str) -> str:
    """Search documentation for relevant information."""
    docs = retriever.invoke(query)
    return "\n\n".join([d.page_content for d in docs])

agent = create_agent(
    model="gpt-4.1",
    tools=[search_docs],
)

result = agent.invoke({
    "messages": [{"role": "user", "content": "How do I create an agent?"}]
})

Create an agent that uses RAG as a tool for answering questions.

import { createAgent } from "langchain";
import { tool } from "@langchain/core/tools";
import { z } from "zod";

const searchDocs = tool(
  async (input) => {
    const docs = await retriever.invoke(input.query);
    return docs.map(d => d.pageContent).join("\n\n");
  },
  {
    name: "search_docs",
    description: "Search documentation for relevant information.",
    schema: z.object({ query: z.string() }),
  }
);

const agent = createAgent({
  model: "gpt-4.1",
  tools: [searchDocs],
});

const result = await agent.invoke({
  messages: [{ role: "user", content: "How do I create an agent?" }],
});

What You CAN Configure

Chunk size/overlap
Embedding model
Number of results (k)
Metadata filters
Search algorithms: Similarity, MMR

What You CANNOT Configure

Embedding dimensions (per model)
Mix embeddings from different models in same store

Chunk size 500-1500 is typically good.

# WRONG: Too small (loses context) or too large (hits limits)
splitter = RecursiveCharacterTextSplitter(chunk_size=50)
splitter = RecursiveCharacterTextSplitter(chunk_size=10000)

# CORRECT
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)

Chunk size 500-1500 is typically good.

// WRONG: Too small or too large
const splitter = new RecursiveCharacterTextSplitter({ chunkSize: 50 });

// CORRECT
const splitter = new RecursiveCharacterTextSplitter({ chunkSize: 1000, chunkOverlap: 200 });

Use overlap (10-20% of chunk size) to maintain context at boundaries.

# WRONG: No overlap - context breaks at boundaries
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=0)

# CORRECT: 10-20% overlap
splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)

Use persistent vector store instead of in-memory to avoid data loss.

# WRONG: InMemory - lost on restart
vectorstore = InMemoryVectorStore.from_documents(docs, embeddings)

# CORRECT
vectorstore = Chroma.from_documents(docs, embeddings, persist_directory="./chroma_db")

Use persistent vector store instead of in-memory to avoid data loss.

// WRONG: Memory - lost on restart
const vectorstore = await MemoryVectorStore.fromDocuments(docs, embeddings);

// CORRECT
const vectorstore = await Chroma.fromDocuments(docs, embeddings, { collectionName: "my-collection" });

Use the same embedding model for indexing and querying.

# WRONG: Different embeddings for index and query - incompatible!
vectorstore = Chroma.from_documents(docs, OpenAIEmbeddings(model="text-embedding-3-small"))
retriever = vectorstore.as_retriever(embeddings=OpenAIEmbeddings(model="text-embedding-3-large"))

# CORRECT: Same model
embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
vectorstore = Chroma.from_documents(docs, embeddings)
retriever = vectorstore.as_retriever()  # Uses same embeddings

Use the same embedding model for indexing and querying.

const embeddings = new OpenAIEmbeddings({ model: "text-embedding-3-small" });
const vectorstore = await Chroma.fromDocuments(docs, embeddings);
const retriever = vectorstore.asRetriever();  // Uses same embeddings

Only opt in to FAISS deserialization for trusted local indexes. Python FAISS indexes include pickle-backed metadata, and untrusted pickle files can execute arbitrary code during loading.

# WRONG: Loading a downloaded, shared, cloud-hosted, or third-party-controlled
# FAISS index with dangerous deserialization enabled.
loaded_store = FAISS.load_local(
    "./untrusted_faiss_index",
    embeddings,
    allow_dangerous_deserialization=True,
)

# CORRECT: Only opt in when the index directory was created by you and has
# remained under your control.
loaded_store = FAISS.load_local(
    "./faiss_index",
    embeddings,
    allow_dangerous_deserialization=True,
)

If you cannot guarantee the provenance of a persisted index, do not load it with allow_dangerous_deserialization=True. Rebuild the index from trusted source documents or use a vector store/backend that does not require pickle deserialization for untrusted files.

Ensure embedding dimensions match the vector store index dimensions.

# WRONG: Index has 1536 dimensions but using 512-dim embeddings
pc.create_index(name="idx", dimension=1536, metric="cosine")
vectorstore = PineconeVectorStore.from_documents(
    docs, OpenAIEmbeddings(model="text-embedding-3-small", dimensions=512), index=pc.Index("idx")
)  # Error: dimension mismatch!

# CORRECT: Match dimensions
embeddings = OpenAIEmbeddings()  # Default 1536

This skill ships inside the LangChain Agent-Building Skills plugin.

Install just this skill

Run this outside your AI tool, using Node.js 22.20.0 or newer. Skills CLI 1.7.2 skips skill selection; prompts cancel installation without an interactive terminal. If choosing Project when asked about installation scope, open your project's folder before running this.

Install checklist

Check selected tools if a picker appears; Claude Code may be unchecked under Additional agents. ↑/↓ moves, Space toggles selections, Enter confirms.
Without a tool picker, check Installing to: for your intended tool. Stop if it names a different tool.
If asked, choose Symlink (Recommended) for shared files or Copy to all agents for separate copies; review the summary before confirming installation.
Install the find-skills skill? defaults to Yes, adding another skill for all projects. Choose No if you don't want that extra installation.

Skills CLI 1.7.2 sends install data to skills.sh by default, including the source, skill names, target tools and file details; GitHub sources are reported only when GitHub confirms the repository is public, while other remote sources can also be reported. Confirmed-public GitHub sources also trigger a security-audit request. Before installing, set the environment variable DO_NOT_TRACK=1 or DISABLE_TELEMETRY=1 to disable both.

npx skills add langchain-ai/langchain-skills --skill "langchain-rag" --full-depth

After installing for Claude Code, open it in the installation project if you chose Project; Global makes the skill available across projects on this computer. Once Claude Code loads the skill, its description helps Claude decide when to use it.

If installation created a top-level skills folder that did not exist when your Claude Code session started, run /reload-skills in that session. Run it again after later changes in that folder.

Install the whole plugin

This route installs the LangChain Agent-Building Skills plugin package selected for your AI tool.

Claude Code

Add LangChain's marketplace, then install the plugin

Type both lines inside a Claude Code session, not at your shell prompt. The first adds LangChain's marketplace, which is this repository; the second opens the details of the langchain-skills plugin, and Claude Code installs it once you pick one of the install options shown. Afterwards, LangChain's README asks you to set OPENAI_API_KEY or ANTHROPIC_API_KEY, for whichever of those model providers you build with.

/plugin marketplace add langchain-ai/langchain-skills

/plugin install langchain-skills@langchain-skills

Open plugin page ↗

What's inside

Files included with this skill.

View on GitHub