RAGLite is a local-first RAG cache that distills documents into structured Markdown for efficient hybrid indexing and retrieval.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install raglite
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install raglite using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is raglite?
RAGLite serves as a robust, local-first retrieval-augmented generation (RAG) cache designed to store and manage information the model wasn't trained on. By leveraging Openclaw Skills, it allows developers to keep sensitive data on their own machines while providing a durable place for local and private knowledge. The skill focuses on a compression-before-embeddings approach, distilling documents into human-readable Markdown first to ensure more reliable retrieval and cheaper prompts.
Instead of relying on managed vector databases, RAGLite uses established open-source tools like Chroma and ripgrep. This architecture provides an auditable and version-controllable system for private notes, medical records, or internal runbooks. It is specifically built to treat source data as untrusted, incorporating security measures against prompt injection during the distillation process.
raglite Use Cases
- Storing and querying local or private knowledge like school work and personal notes.
- Managing sensitive medical records or internal company runbooks with local-first privacy.
- Reducing model context fluff by distilling large documents into structured Markdown.
- Performing repeated lookups on data that is not part of the model's original training set.
How raglite Works
- The distillation engine, which defaults to OpenClaw, processes source documents into structured Markdown to remove duplication.
- The system indexes the distilled content into a local Chroma vector database.
- Ripgrep is utilized alongside the vector database to enable powerful hybrid retrieval (keyword + vector search).
- Users query the collection, and RAGLite retrieves the most relevant context from the local cache to inform the agent's response.
raglite Setup
To get started with RAGLite for your Openclaw Skills setup, ensure you have python3, pip, and ripgrep (rg) installed. Run the following command to create a local virtual environment and install necessary packages:
./scripts/install.sh
This installs the raglite-chromadb package into a skill-local directory.
raglite Data Schema & Taxonomy
RAGLite organizes its artifacts to ensure they are both machine-readable and human-auditable. The data structure is defined as follows:
| Component | Type | Description |
|---|---|---|
| Output Directory | Filesystem Path | Stores distilled Markdown files and the local Chroma database. |
| Distilled Docs | Markdown | Human-readable, compressed versions of the source material. |
| Vector Index | Chroma DB | Local vector storage used for semantic search lookups. |
| Engine | Configuration | Specifies the processing engine, defaulting to OpenClaw. |
raglite Advanced Features
- Hybrid retrieval pipeline combining vector-based semantic search with ripgrep keyword matching.
- Compression-before-embeddings strategy to maximize retrieval accuracy and minimize token usage.
- Integrated prompt injection mitigation to handle untrusted third-party source data safely.
- Flexible collection management allowing multiple isolated knowledge bases within the same Openclaw Skills environment.
SKILL.md
Loading
Related Openclaw Skills
RAGLite
A local-first RAG cache that distills documents into structured Markdown and indexes them for hybrid vector and keyword search.
raglite
A local-first RAG cache that distills documents into structured Markdown and indexes them for hybrid vector and keyword retrieval.
DeepthinkLite
DeepthinkLite is a local-first research tool that generates durable markdown artifacts and enforces a time-budgeted workflow for deep thinking.
Anakin Web Data Extraction
A powerful CLI tool for converting websites into clean, structured data using AI-driven scraping and autonomous research agents.
RAGLite
A local-first RAG cache that distills documents into structured Markdown for high-efficiency hybrid indexing and retrieval.
Nimrobo CLI
A comprehensive command-line interface for managing AI-driven voice interviews and professional matching networks.