A high-performance local Retrieval-Augmented Generation suite for efficient code and document search within local workspaces.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install local-file-rag-basic
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 local-file-rag-basic 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 Local File RAG Search (Basic Edition)?
Local File RAG Search (Basic Edition) is a specialized toolset designed to give AI agents deep context into local projects. By integrating this into your Openclaw Skills library, you allow your agent to intelligently scan, index, and retrieve information from a variety of file types including JS/TS, Python, C++, Go, Markdown, PDF, DOCX, and XLSX.
This version is optimized for resource efficiency, utilizing single-threaded sequential indexing to ensure low overhead on your system. It is specifically engineered for developers who need reliable code retrieval and document analysis without the heavy hardware requirements of full-scale vector databases.
Local File RAG Search (Basic Edition) Use Cases
- Finding specific function definitions or logic patterns across multiple source code files.
- Extracting technical details from local PDF or DOCX documentation to help solve coding tasks.
- Analyzing data patterns within local XLSX spreadsheets using natural language queries.
- Providing high-relevance code snippets to an LLM to reduce token usage and improve response accuracy.
How Local File RAG Search (Basic Edition) Works
- The AI agent initiates a search using the local_file_rag_search tool with a specific query or function name.
- The system identifies files within the specified root directory, filtering out any files exceeding the 20MB safety limit.
- The skill performs a sequential, single-threaded scan to index the content while maintaining low CPU and RAM usage.
- Relevant segments are extracted and structured into an output protocol consisting of file skeletons, metadata, and clustered code snippets.
- The structured results are passed back to the agent to provide grounded, factual context for its final output.
Local File RAG Search (Basic Edition) Setup
To integrate this capability into your environment, ensure your workspace is configured to recognize Openclaw Skills.
# Download or clone the local-file-rag-basic skill into your project
# No external heavy-weight dependencies are required for the Basic Edition
Simply point your agent configuration to the directory containing the skill metadata to begin searching local files.
Local File RAG Search (Basic Edition) Data Schema & Taxonomy
The skill organizes and returns data based on a strict internal protocol to ensure the AI agent receives the most relevant context:
| Component | Description |
|---|---|
| Skeletons | A structural outline of the file (e.g., class names, function headers) |
| Metadata | Information regarding file path, size, and indexing status |
| Clustered Snippets | Grouped sections of code or text that most closely match the search query |
All indexed files must remain under the 20MB threshold to ensure performance stability.
Local File RAG Search (Basic Edition) Advanced Features
- Multi-format compatibility spanning from high-level programming languages to office productivity documents.
- Targeted search capabilities allowing for restricted scanning of specific directories or single files.
- Resource-optimized indexing engine designed for standard developer workstations.
- Seamless integration with the broader ecosystem of Openclaw Skills for multi-agent workflows.
SKILL.md
Loading
Related Openclaw Skills
AI Frens Onboarding
Transform your OpenClaw agent into a portable AI Fren with its own tokenized economy, treasury, and cross-platform payment capabilities.
AI Frens Ambassador
A guide and workflow for AI agents to advocate for economic autonomy and self-funding infrastructure through the AI Frens ecosystem.
WhatsApp GIF Integration for AI Agents
A specialized skill for AI agents to find and deliver contextually relevant GIFs directly to WhatsApp conversations.
Questrade Trading Automation Skill
A secure browser-based trading workflow for Questrade with integrated Yahoo Finance verification and automated safety checklists.
ZeeLin Claw Swarm
A multi-group communication platform that enables AI agents to read, monitor, and participate in topical chat discussions via a structured REST API.
US-Iran War Situation Tracker
A high-frequency tactical intelligence skill designed to monitor the US-Iran conflict using a strict 6-hour reporting window and multi-source triangulation.
Featured*
