An intelligent tool that launches parallel agents to conduct deep, multi-topic research and generate structured knowledge bases.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install multi-search
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 multi-search 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 Multi-Topic Deep Research?
The Multi-Topic Deep Research skill is a high-performance engine designed to integrate seamlessly with Openclaw Skills. It automates the complex task of analyzing source materials and generating systematic documentation by deploying multiple independent research agents in parallel. This skill is built on core principles of factual integrity and professional synthesis, ensuring that information retrieval is deep, concise, and focused on solving specific problems without fabricating data.
By leveraging independent task execution, this skill transforms raw materials or topical prompts into a concise research knowledge base. It is universally applicable across legal, business, technical, and academic fields, making it a versatile addition to any developer or researcher's toolkit. The systematic approach ensures that every research thread is exhaustive yet avoids redundant information through a sophisticated deduplication mechanism.
Multi-Topic Deep Research Use Cases
- Conducting deep, simultaneous research on multiple related topics to save time.
- Systematic information retrieval from large volumes of raw input materials or documents.
- Generating structured research reports that require multi-perspective analysis integration.
- Building a professional knowledge base for complex technical, legal, or business projects.
How Multi-Topic Deep Research Works
- Analysis Preparation: The skill reads input materials and extracts a list of unique research topics, ensuring each has a clear direction.
- Directory Selection: It automatically detects or creates an organized project structure, typically under a 03 - Deep Research folder.
- Parallel Agent Launch: An independent research agent is assigned to each topic, receiving specific context, objectives, and retrieval scopes.
- Deduplication Protocol: Each agent declares its intended search keywords; the main agent reviews these to prevent overlapping efforts and ensures unique angles.
- Deep Retrieval: Agents perform 4-6 rounds of deep searching using WebSearch and WebFetch tools to gather comprehensive content.
- Integration & Output: The skill synthesizes findings into an overview document and individual topic reports with inter-document navigation and actionable recommendations.
Multi-Topic Deep Research Setup
This skill utilizes built-in capabilities within the Openclaw Skills environment and requires no complex external configuration. You can trigger the research process using the following bash commands:
# Method 1: Using a document as source
/multi-search @your-document.md
# Method 2: Specifying topics directly
/multi-search
Project: [Project Name]
Research Topics:
1. [Topic One]
2. [Topic Two]
Multi-Topic Deep Research Data Schema & Taxonomy
The skill organizes data into a hierarchical structure to ensure clarity and easy navigation. All files are stored within the project sub-directory:
| Component | Path/Pattern | Description |
|---|---|---|
| Research Folder | [project-name]/03 - Deep Research/ |
The primary container for all generated research assets. |
| Overview Doc | 000.Research-Overview.md |
A high-level summary containing core findings and action items. |
| Topic Reports | YYMMDD [Topic Name].md |
Detailed reports for each specific research thread. |
| Numbering | 00. to 20+ |
A systematic numbering convention for overview, core, and extended research. |
Multi-Topic Deep Research Advanced Features
- Parallel Agent Execution: Scales research by running multiple independent agents simultaneously.
- Dynamic Keyword Adjustment: Agents pivot their search strategy in real-time if the main agent identifies redundant coverage.
- Contextual Integrity: Transfers project background and core problems to sub-agents to maintain research relevance.
- Inline Citation Standard: Automatically embeds source links (web, academic, institutional) directly within the narrative text for immediate verification.
- Actionable Synthesis: Generates immediate action lists and strategic recommendations based on integrated findings.
SKILL.md
Loading
Related Openclaw Skills
Feishu Multi-Account Management
A powerful configuration skill that enables OpenClaw to route specific Feishu bot accounts to dedicated AI sub-agents.
Image Reader
A sophisticated multimodal tool designed to extract text and generate detailed descriptions from images using advanced AI models.
DNS Robot
A comprehensive toolkit for performing DNS queries, SSL audits, and email security checks without requiring an API key.
Live Shopping Notes
A specialized AI skill for capturing, tagging, and summarizing product offers and details during live commerce events.
WeCom AI Bot Setup
A streamlined utility to connect OpenClaw to Enterprise WeChat (WeCom) using the AI Bot WebSocket mode for seamless internal team automation.
UI Inspiration Library
A structured visual search system that archives UI screenshots and retrieves them based on style, page type, and use case.
Featured*
