Research Agent for Openclaw

A sophisticated AI research specialist that transforms vague inquiries into comprehensive, source-backed analysis reports and comparison matrices.

jahonn
v1.0.0
Mar 21, 2026
1
2k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install deep-research-agent

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install deep-research-agent 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 Research Agent?

Research Agent is a robust framework designed to automate the heavy lifting of information gathering and synthesis. By leveraging Openclaw Skills, it transitions from simple web lookups to complex investigation methodologies, ensuring developers and researchers receive high-utility data instead of generic summaries. Whether you are performing a quick fact check or a multi-day market analysis, this skill structures the output into actionable insights.

The agent employs a tiered logic system, selecting the most appropriate LLM—such as Haiku for speed or Sonnet for deep reasoning—based on the complexity of the research task. It excels at identifying patterns across disparate sources, evaluating the credibility of documentation versus community sentiment, and providing objective recommendations for technical decision-making.

Research Agent Use Cases

  • Investigating emerging technology trends and landscape analysis.
  • Conducting side-by-side competitor or tool comparisons (e.g., Cursor vs. VS Code).
  • Performing market research and identifying gaps in existing product categories.
  • Evaluating software frameworks or libraries for long-term project adoption.
  • Generating executive summaries and deep-dive RESEARCH.md reports for stakeholders.

How Research Agent Works

  1. Trigger Identification: The agent detects intent through keywords like research, compare, or investigate to determine the required mode.
  2. Strategy Selection: Based on the prompt, it selects one of five modes: Quick, Deep Dive, Compare, Landscape, or Evaluate.
  3. Information Gathering: It executes multi-source searches across official documentation, technical blogs, and community forums like Reddit or Hacker News.
  4. Critical Evaluation: The skill filters results based on recency, credibility, and potential bias to ensure data integrity.
  5. Synthesis & Reporting: Findings are organized into a structured format, such as a comparison matrix or a detailed Markdown report with verified URLs.

Research Agent Setup

To integrate Research Agent into your local environment using Openclaw Skills, use the following commands:

# Navigate to your agent skills directory
cd ~/openclaw/skills

# Install the research-agent skill
openclaw install research-agent

# Verify installation
openclaw list

Research Agent Data Schema & Taxonomy

The Research Agent organizes output data into specific files and formats based on the depth of the inquiry:

File Name Format Content Description
RESEARCH.md Markdown Comprehensive deep-dive report including executive summary, key findings, and full source list.
LANDSCAPE.md Markdown Categorized map of direct and adjacent competitors, including gap analysis.
Stdout Plain Text Comparison matrices, quick summaries, and decision scores for immediate feedback.
Metadata YAML/Frontmatter Tracks research duration, model used, and confidence levels.

Research Agent Advanced Features

  • Multi-model routing utilizing Haiku for rapid queries and Sonnet for complex reasoning and judgment.
  • Automated subagent spawning for deep-dive tasks to prevent main context exhaustion.
  • Custom decision frameworks for technical evaluation using weighted RICE or ICE scoring.
  • Pattern extraction algorithms that identify points of consensus and disagreement across multiple web sources.
  • Source-verified citations with automated URL verification for academic and technical rigor within Openclaw Skills.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*