A professional-grade research agent skill that utilizes Model Context Protocol (MCP) tools to perform deep web analysis and generate cited technical reports.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deep-research-with-mcp
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-with-mcp 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 with MCP?
Deep Research with MCP is a sophisticated Openclaw Skills module designed for autonomous agents that need to go beyond simple LLM knowledge cutoffs. By leveraging the Model Context Protocol, it connects your agent to live web search engines like MiniMax and zai-mcp-server, allowing for real-time information retrieval.
This skill transforms a standard AI interaction into a rigorous investigative process. It doesn't just summarize snippets; it performs multi-step sub-question planning, deep-reads source content via curl, and cross-references data to ensure high-fidelity reporting with verifiable citations.
Deep Research with MCP Use Cases
- Conducting deep dives into emerging technologies or market trends with real-time data.
- Comparing technical frameworks, such as Rust vs Go, with the latest 2026 benchmarks.
- Generating comprehensive executive summaries for business strategy and bootstrapping.
- Investigating complex regulatory or academic topics requiring multi-source verification.
How Deep Research with MCP Works
- Verify MCP search tool availability using mcporter to ensure MiniMax or web-search-prime is active.
- Define research goals and break the primary topic into 3-5 specific sub-questions for maximum coverage.
- Execute multi-source searches using MCP tools, targeting 15-30 unique reputable sources.
- Perform deep-reading by fetching full HTML content of primary sources to extract detailed insights beyond search snippets.
- Synthesize findings into a structured Markdown report featuring executive summaries, inline citations, and a full bibliography.
- Save the final report to the designated research directory and present key takeaways to the user.
Deep Research with MCP Setup
To use this within the Openclaw Skills ecosystem, ensure you have the required binaries and configured MCP servers:
# Install dependencies
# Ensure mcporter, curl, and python3 are in your PATH
# List and verify your MCP search tools
mcporter list
# Example: Calling a search manually to test configuration
mcporter call MiniMax.web_search query:"Openclaw Skills documentation"
Deep Research with MCP Data Schema & Taxonomy
The skill maintains a clean organizational structure for all research projects to ensure data persistence and easy retrieval:
| File/Folder | Description |
|---|---|
research/[slug]/ |
The primary directory for a specific research task. |
report.md |
The final synthesized report containing findings and citations. |
methodology |
A section within the report detailing queries used and sources analyzed. |
confidence_score |
A metadata tag (High/Medium/Low) based on source consistency. |
Deep Research with MCP Advanced Features
- Multi-agent support: Can be spawned as a sub-agent task using sessions_spawn for asynchronous background research.
- Maker-Checker verification: Implements quality rules that flag unverified claims and prioritize recent data (last 12 months).
- Automated Content Extraction: Uses a Python-based pipeline to strip HTML and process up to 5,000 characters of raw text per source.
- Citation Engine: Automatically generates inline citations and a full source list with summaries.
SKILL.md
Loading
Related Openclaw Skills
Linkup Search
An agentic web search and retrieval skill that enables AI agents to find, scrape, and synthesize real-time internet data through the Linkup API.
Academic Literature Search Skill
A professional-grade academic research tool that integrates multiple authoritative databases for comprehensive, high-speed literature retrieval and analysis.
Link Building Outreach
A sophisticated automation pipeline designed to scale personalized link-building outreach without sacrificing quality.
Tavily Search Skill
A high-performance web search skill using the Tavily API to provide AI agents with real-time, filtered internet access and source discovery.
Threads Downloader
An automated AI agent skill designed to parse Threads links and download high-definition video content directly to your local storage.
UAPI Daily News Image
An efficient AI agent skill designed to fetch and generate daily news images, posters, and digests through a single optimized UAPI endpoint.
Featured*
