A self-contained research agent that performs multi-source web searches and synthesizes professional, cited reports without requiring external API keys.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deep-research-pro
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 deep-research-pro 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 Deep Research Pro?
Deep Research Pro is a specialized tool designed for Openclaw Skills that transforms the way agents handle information gathering. By utilizing DuckDuckGo for both web and news searches, it bypasses the need for expensive search API keys while maintaining high-quality output. It functions as an end-to-end research assistant that can autonomously break down complex topics into sub-questions, extract full-text content from various URLs, and cross-reference data to ensure accuracy.
This skill is particularly valuable for users who need more than just a surface-level summary. It dives deep into academic, official, and reputable news sources to provide a synthesis that includes confidence levels, source methodology, and actionable insights. By integrating Deep Research Pro into Openclaw Skills, developers can empower their agents to produce high-fidelity whitepapers and market analysis reports on demand.
Deep Research Pro Use Cases
- Investigating market trends and competitor landscapes for business strategy.
- Conducting deep dives into technical frameworks like Rust vs Go for engineering decisions.
- Monitoring current events and emerging technologies through multi-source news synthesis.
- Bootstrapping comprehensive research reports for academic or professional documentation.
How Deep Research Pro Works
- The agent begins by clarifying the user's specific goals, depth requirements, and intended use for the research.
- The topic is decomposed into 3-5 logical sub-questions to ensure the research covers all necessary angles.
- The skill executes multi-source searches using DuckDuckGo, targeting both general web results and recent news articles.
- It fetches full-page text from the most relevant sources using curl and a Python-based extraction script to move beyond simple search snippets.
- The findings are synthesized into a structured Markdown report featuring an executive summary, themed sections, and inline citations.
- The final output is saved locally to the research directory for persistent storage and easy delivery to the user.
Deep Research Pro Setup
To get started with Deep Research Pro within Openclaw Skills, ensure your environment has the necessary dependencies installed.
# Ensure the DuckDuckGo search skill is accessible
ls /home/clawdbot/clawd/skills/ddg-search/scripts/ddg
# Verify Python 3 and curl are available for text extraction
python3 --version
curl --version
# Create the research directory for report storage
mkdir -p ~/clawd/research/
Deep Research Pro Data Schema & Taxonomy
Deep Research Pro maintains a structured data hierarchy for all generated findings. All research projects are organized by a unique slug identifier.
| File Path | Description |
|---|---|
~/clawd/research/[slug]/report.md |
The primary research report containing all findings and citations. |
~/clawd/research/[slug]/sources.txt |
(Optional) A raw list of all URLs analyzed during the process. |
Each report generated by this skill for Openclaw Skills includes metadata such as the generation date, number of unique sources analyzed, and a confidence score based on the consistency of the findings.
Deep Research Pro Advanced Features
- Sub-agent spawning support for background processing of intensive, long-form research tasks.
- Automated text parsing that strips HTML tags to provide clean, readable context for the LLM.
- Multi-query search strategy using keyword variations to maximize source diversity.
- Cross-referencing logic that flags unverified claims appearing in only a single source.
- Capability to handle 15-30 unique sources per report for maximum depth.
SKILL.md
Loading
Related Openclaw Skills
Wash Trade Detector
An automated detection engine for identifying artificial NFT transaction volume and price manipulation through 7 confidence-weighted patterns.
Data-Normalization (Artledger)
A specialized data integrity skill that standardizes raw records from multiple sources to ensure accurate identity resolution and enrichment.
Confidence Scoring Gatekeeper
A mandatory data integrity gatekeeper that validates and enforces confidence scoring rules before writing entity links to production databases.
webserp
A powerful metasearch CLI that enables AI agents to query multiple search engines in parallel without requiring API keys.
WidgetDesk macOS Widget Manager
A professional toolset for managing Übersicht desktop widgets on macOS through automated scripts and JSX templates.
Mindmap Generator
A professional AI tool that converts complex conversations, decisions, and project plans into visual mindmap PNGs delivered instantly to Telegram.
Featured*
