A high-fidelity research agent that orchestrates parallel subagents to perform deep-dive strategic intelligence and cross-source validation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deep-research-surf
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-surf 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 (Surf)?
The Deep Research (Surf) skill is a sophisticated intelligence engine designed to move beyond simple search queries. It utilizes the Surf MCP suite to perform comprehensive, multi-layered investigations across diverse platforms including GitHub, Reddit, Twitter, and YouTube. By deploying parallel subagents, it ensures that every topic is analyzed from broad, technical, and contrarian perspectives simultaneously.
This skill is built for professionals who require synthesis over simple summarization. It focuses on evidence-rich findings, cross-source validation, and identifying orthogonal insights—those hidden gems that single-pass searches often miss. It transforms raw web data into actionable strategic reports with a focus on technical accuracy and causal relationships.
Deep Research (Surf) Use Cases
- Competitive landscape analysis to identify market white space and competitor trajectories.
- Technical deep dives into architecture patterns, engineering designs, and repository audits.
- Strategic intelligence gathering for executive decision-making and product planning.
- Sentiment analysis and community feedback extraction from Reddit, X, and specialized forums.
- Trend forecasting and novelty seeking to identify emerging signals and contrarian views in specific industries.
How Deep Research (Surf) Works
- Topic Assessment: The agent restates the query to ensure clarity and plans which Surf primitives (web, GitHub, Reddit, etc.) are most relevant.
- Multi-Angle Discovery: Three parallel subagents are spawned to gather broad overviews, diversified technical/user perspectives, and emerging trends.
- Source Selection & Deep Dive: Based on discovery findings, the agent identifies high-signal sources for full-content extraction using specialized crawl tools.
- Coverage Validation: A gap-detection checklist is run to ensure metrics, leadership views, and technical details are all present.
- Cross-Source Analysis: The agent triangulates claims across independent sources, detects contradictions, and extracts unique insights.
- Synthesis & Reporting: A structured Markdown report is generated, highlighting key findings, strategic recommendations, and annotated sources.
Deep Research (Surf) Setup
To enable this skill within your environment, ensure the Surf MCP tools are configured and then invoke the skill using the Openclaw Skills command structure:
# Trigger a deep research session via CLI
/deep-research-surf [your topic here]
Ensure your agent has permissions to spawn background tasks for the parallel subagent execution phase.
Deep Research (Surf) Data Schema & Taxonomy
The skill organizes research data into a hierarchical corpus before final synthesis:
| Data Layer | Content Type | Metadata Tracked |
|---|---|---|
| Discovery Corpus | Broad themes and initial URLs | Source reliability and relevance scores |
| Deep Dive Extraction | Full text, transcripts, and READMEs | Timestamps, author credibility, and quotes |
| Validation Matrix | Cross-referenced facts and metrics | Confidence levels and source count |
| Final Synthesis | Thematic insights and strategic advice | Clickable Markdown links and annotations |
Deep Research (Surf) Advanced Features
- Parallel Subagent Orchestration: Simultaneously runs multiple general-purpose agents to reduce research latency.
- Context Budget Engineering: Automatically adjusts context windows based on whether it is performing broad scanning or deep content extraction.
- Novelty-Seeking Algorithm: Explicitly programmed to find contrarian opinions and lesser-known strategies to avoid echo chambers.
- Cross-Platform Primitives: Native support for extracting YouTube subtitles, GitHub commit history, and social media threads for a 360-degree view.
SKILL.md
Loading
Related Openclaw Skills
last30days-surf
A parallel research engine that synthesizes 30 days of cross-platform social engagement and web data into actionable briefs.
AMC Entertainment & Networks Analysis
A comprehensive analytical skill for evaluating the business transformation of the world's largest cinema chain and its premium content network.
DS Automobiles Research Skill
An AI-powered research module for analyzing the business strategy, historical heritage, and luxury positioning of DS Automobiles.
10x Genomics
A revolutionary single-cell genomics platform transitioning biological research from population averages to individual cell analysis.
Meta Ads CLI
A secure command-line interface skill for managing Meta advertising campaigns, ad sets, and insights via an AI-powered agent.
HP Hewlett Packard Knowledge Skill
A specialized intelligence module for analyzing Hewlett-Packard's corporate evolution and hardware market dominance.
Featured*
