chonkie
🦛 CHONK docs with Chonkie ✨ — The lightweight ingestion library for fast, efficient and robust RAG pipelines
Chonkie is an AI agent skill that provides coding assistants with comprehensive knowledge of the Chonkie text chunking library for building RAG (Retrieval-Augmented Generation) pipelines. The skill enables agents like Claude Code, Cursor, and Copilot to help developers select appropriate chunkers, configure pipelines, choose tokenizers, set up embeddings, and integrate with vector databases. It covers all chunking strategies (semantic, recursive, token-based, code-aware), the Pipeline API, async/batch processing patterns, and the REST API server. Developers install the skill once and their AI assistant gains expert-level understanding of Chonkie's entire ecosystem.
Key Features
Use Cases
- 01Getting AI assistant help to select the right chunker for a specific RAG use case (semantic vs recursive vs code-aware)
- 02Building end-to-end RAG pipelines with agent guidance on fetchers, chunkers, embeddings, and vector store integration
- 03Configuring async chunking workflows for high-throughput document processing applications
- 04Setting up the Chonkie REST API server with proper dependencies and pipeline definitions
- 05Optimizing chunk sizes and overlap strategies for different embedding models and retrieval scenarios
- 06Integrating Chonkie with specific vector databases like Pinecone, Qdrant, or ChromaDB using handshakes
Related Skills
View moresuperpowers
An agentic skills framework & software development methodology that works.
skills
Skills for Real Engineers. Straight from my .agents directory.
skills
Public repository for Agent Skills
ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
chonkie — FAQ
What is the Chonkie agent skill?+
The Chonkie agent skill is a knowledge plugin that teaches AI coding assistants (Claude Code, Cursor, Copilot, etc.) about the Chonkie text chunking library, including all chunkers, pipelines, embeddings, and vector database integrations. It enables your AI assistant to help you build RAG pipelines using Chonkie's API.
How do I install the Chonkie skill for my AI agent?+
For most agents, run 'npx skills add chonkie-inc/skills' using skills.sh. For Claude Code specifically, use '/plugin marketplace add chonkie-inc/skills' followed by '/plugin install chonkie' in the chat interface.
Which AI coding agents support the Chonkie skill?+
The skill works with Claude Code, Cursor, GitHub Copilot, Gemini CLI, and 20+ other coding agents that support the skills.sh format or Claude Code plugin marketplace.
Do I need API keys or the Chonkie library installed to use the skill?+
The skill itself only teaches your agent about Chonkie's API—no keys required for the skill. However, to actually run Chonkie code that the agent helps you write, you'll need to install the Chonkie Python library and may need API keys for embedding providers or LLMs you choose to use.
Is the Chonkie agent skill free to use?+
Yes, the Chonkie agent skill is free and open source under the same license as the Chonkie library. The underlying Chonkie library itself is also free, though some optional integrations (OpenAI, Cohere, etc.) require paid API keys.
How do I install chonkie?+
Open the source repository on GitHub and follow its README. chonkie is a skill — MCP Agents Market links you directly to the official repo.
Is chonkie free?+
chonkie is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.