Chroma

Direct export to Chroma vector database—perfect for local development.

Quick Start

skill-seekers scrape --format chroma --config configs/react.json

Features

  • ✅ Local vector storage - No cloud required
  • ✅ Embeddings included - Uses default embedding model
  • ✅ Metadata filtering - Filter by category, source, language
  • ✅ Persistent storage - Data saved between sessions

Python Example

import chromadb

# Connect to Chroma
client = chromadb.PersistentClient(path="./chroma_db")

# Get collection
collection = client.get_collection("react-docs")

# Query
results = collection.query(
    query_texts=["How do I use useState?"],
    n_results=3
)

print(results['documents'][0])

With LangChain

from langchain_community.vectorstores import Chroma
from langchain_openai import OpenAIEmbeddings

vectorstore = Chroma(
    collection_name="react-docs",
    embedding_function=OpenAIEmbeddings(),
    persist_directory="./chroma_db"
)

results = vectorstore.similarity_search("React Hooks")

Next Steps