ResearchMate is a professional research skill that automates high-fidelity material collection with a rigorous four-fold anti-hallucination verification process.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install research-mate
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 research-mate 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 ResearchMate?
ResearchMate is a specialized tool designed for users of Openclaw Skills who require deep, systematic material collection for professional writing. It functions as a digital research assistant that moves beyond simple search queries by employing a structured three-step interaction to define research scope, followed by multi-source routing. This skill is particularly valuable for long-form content creators who need to ensure their articles, reports, or scripts are backed by verified data and credible sources.
By integrating directly into the AI agent workflow, ResearchMate ensures that the content gathered for Openclaw Skills is not just relevant, but also accurate. It targets high-quality sources such as financial reports, technical media, and official government announcements, providing a structured material library that drastically reduces the manual effort of fact-checking and information organization.
ResearchMate Use Cases
- Collecting verified data and case studies for deep-dive technical articles.
- Performing competitive research and financial analysis for business reports.
- Gathering multi-source materials for high-quality video scripts and media content.
- Conducting systematic literature or policy reviews for academic and professional projects.
How ResearchMate Works
- Interactive Requirement Analysis: The skill triggers a mandatory three-step query to identify the topic, key entities (companies, products, or people), and the intended word count.
- Intelligent Source Routing: Based on the topic, ResearchMate automatically selects the best source combinations, such as tech media, financial terminals, or official government bulletins.
- Massive Data Extraction: It calculates and gathers approximately 8 to 10 times the volume of the target content to provide a deep pool of reference material.
- Four-Fold Verification: Every piece of data is checked for numerical accuracy, subject clarity, timestamp validity, and source reliability to prevent hallucinations.
- Quality Grading and Scoring: Materials are weighted and scored (0-100) across four dimensions: credibility, timeliness, completeness, and cross-verification.
- Structured Output Generation: The skill delivers a formatted Markdown material library and a CSV evaluation table for easy integration into the final writing process.
ResearchMate Setup
To activate ResearchMate within your environment, ensure your agent has the necessary permissions for web search and file writing. You can initialize the research process by using the following command structure:
# Trigger the material collection workflow
research-mate --topic "Your Research Topic" --keywords "key,terms" --length 3000
This skill is designed to work seamlessly with other Openclaw Skills to provide a robust research-to-writing pipeline.
ResearchMate Data Schema & Taxonomy
ResearchMate organizes its findings into two distinct delivery formats to ensure clarity and usability:
| File Type | Purpose | Key Metadata |
|---|---|---|
| Markdown Library | Primary content repository | Source URL, Publishing Date, Verification Status, Citation Suggestions |
| CSV Assessment Table | Quality control and auditing | Credibility Score, Timeliness, Cross-Verification Grade (S/A/B/C) |
The schema ensures that every data point within the Openclaw Skills database is traceable and graded for reliability.
ResearchMate Advanced Features
- Smart Volume Calculation: Automatically adjusts the depth of research based on the target length of the final output (8-10x rule).
- Automated Anti-Hallucination Guardrails: Prevents the use of vague terms like "recently" or "certain companies" by requiring specific dates and names.
- Quality Grade Filtering: Automatically flags or excludes low-grade (C-tier) information to maintain the integrity of the research.
- Multi-Dimensional Weighted Scoring: Uses a custom algorithm to calculate an overall reliability score based on source authority and cross-references.
SKILL.md
Loading
Related Openclaw Skills
Visual Bug Hunter
An AI-driven skill for detecting, diagnosing, and fixing visual bugs in GUI and Web applications using computer vision and automated code mapping.
AgentPulse
An intelligent infrastructure monitoring solution that uses a thin agent and cloud-based AI to learn server baselines and automate remediation.
Snickers Brand Intelligence
A comprehensive intelligence skill for analyzing the Snickers brand, its global market dominance, and iconic advertising strategies.
LAS AI PDF Parser (Doubao)
A high-fidelity AI-powered PDF parser that transforms complex documents into structured Markdown using Volcengine Doubao models.
Model Switcher
A robust utility for safely switching the default AI model in your environment with built-in validation and automated backups.
Web Security Penetration Testing
A comprehensive security toolkit for automated web application vulnerability identification, exploitation, and reporting within the Openclaw ecosystem.
Featured*
