chroma

Embed and search documents with metadata using chromadb and sentence-transformers.

Updated May 11, 2026

Search Tags:#semantic search #vector database #document retrieval #metadata filtering #embedding database

Ready to run

npx skills add https://github.com/richardnguyen0715/keep-it-real --skill chroma-richardnguyen0715

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: chroma
Source: https://github.com/richardnguyen0715/keep-it-real/tree/main/refer-projects/hermes-agent/optional-skills/mlops/chroma
Command: npx skills add https://github.com/richardnguyen0715/keep-it-real --skill chroma-richardnguyen0715

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires chromadb, sentence-transformers, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the challenge of efficiently managing and retrieving large document collections, providing fast and accurate semantic search capabilities for AI applications.

Core Features & Use Cases

Vector Database: Store embeddings and metadata, perform vector and full-text search.
Metadata Filtering: Filter by metadata fields for precise document retrieval.
Use Case: Ideal for building RAG (retrieval-augmented generation) applications, semantic search over documents, and local/self-hosted vector databases.

Quick Start

Use the chroma skill to create a new collection and add documents with metadata, then perform a search query.

Frequently Asked Questions about chroma

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I implement semantic search over documents for a RAG application?▼

Semantic search for RAG applications uses vector representations of documents stored in a database to retrieve relevant context. This Skill provides the embedding generation, metadata filtering, and retrieval pipeline needed to build this workflow.

Can I filter document retrieval by metadata fields in a vector database?▼

Yes, metadata filtering is supported. You can attach metadata fields to documents when adding them to the vector database, then apply precise filtering conditions during search queries to restrict the retrieved results.

Do I need sentence-transformers to use an embedding database?▼

Yes, sentence-transformers is required to generate vector embeddings from text. The chromadb dependency manages database storage and retrieval, while sentence-transformers handles converting text into searchable vector representations.

What is the best way to set up a local vector database for AI applications?▼

Setting up a local vector database involves creating a collection to store document embeddings and metadata. This Skill provides a self-hosted solution using chromadb, enabling efficient document retrieval without external cloud services.

Does chromadb support full-text search in addition to vector search?▼

Yes, chromadb supports both vector search and full-text search. This allows semantic similarity matching based on embeddings alongside traditional keyword-based searches within the same document collection.