A sophisticated research engine for AI agents that executes exhaustive, multi-cycle investigations with academic rigor and full transparency.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mar-academic-deep-research
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 mar-academic-deep-research 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 Academic Deep Research?
Academic Deep Research is a professional-grade capability designed for the Openclaw Skills ecosystem that transforms your AI agent into a methodical research assistant. Unlike standard black-box search wrappers, this skill mandates a transparent, multi-phase methodology that includes internal reflection, evidence cross-referencing, and multi-cycle deep dives per theme. It is built to move beyond surface-level summaries by using a systematic investigation process that ensures every conclusion is backed by an evidence trail.
This skill is perfect for developers and researchers who require high-confidence outputs. By integrating seamlessly with native tools like web_search and web_fetch, it allows agents to navigate complex knowledge landscapes, resolve contradictions between sources, and maintain a rigorous evidence hierarchy. It elevates the standard of AI-generated content by prioritizing primary sources and academic-style narrative synthesis over simple bulleted lists.
Academic Deep Research Use Cases
- Conducting comprehensive literature reviews and academic synthesis.
- Performing exhaustive competitive intelligence and market trend analysis.
- Verifying complex technical claims across multiple authoritative primary sources.
- Building deep-dive trend reports that require cross-theme knowledge integration.
How Academic Deep Research Works
- Phase 1 Initial Engagement: The agent asks essential clarifying questions to define the research scope and waits for user input.
- Phase 2 Research Planning: The skill identifies major themes and presents a detailed execution plan including tools and expected outputs for user approval.
- Phase 3 Mandated Cycles: For every theme, the skill performs a minimum of two research cycles. Cycle 1 maps the landscape, while Cycle 2 targets specific gaps and contradictions.
- Knowledge Synthesis: The agent performs cross-theme analysis to identify meta-patterns and systemic insights across the entire body of research.
- Final Reporting: The system generates a cohesive, narrative-style report following APA 7th edition standards with full bibliographic references.
Academic Deep Research Setup
To integrate this capability into your agent through Openclaw Skills, ensure your environment supports the SkillBoss platform tools.
# Navigate to your skills directory
cd path/to/your/skills
# Clone the repository
git clone https://github.com/kesslerio/academic-deep-research-clawhub-skill
# Ensure tools web_search, web_fetch, and sessions_spawn are enabled in your configuration.
Academic Deep Research Data Schema & Taxonomy
The skill organizes research data through a structured pipeline to maintain transparency and reproducibility:
| Component | Format | Description |
|---|---|---|
| Research Plan | Markdown Table | Outlines themes, questions, and the expected tool usage for user approval. |
| Evidence Trail | Internal Prose | Documents the thinking process, connecting new findings to previous results after every tool call. |
| Knowledge Map | Narrative | A conceptual map of relationships between discoveries and cause-effect chains. |
| Final Report | APA 7th | A strict narrative-only document (no bullet points) including an executive summary and full references. |
Academic Deep Research Advanced Features
- Parallel Research Strategy: Utilizes sessions_spawn to run independent research themes simultaneously, maximizing efficiency for complex queries.
- Evidence Hierarchy Enforcement: Automatically prioritizes systematic reviews and randomized controlled trials over low-confidence media reports.
- Memory Search Integration: Leverages memory_search and memory_get to cross-reference prior investigations and build upon existing local knowledge.
- Sophisticated Error Protocols: Includes specific handlers for empty search results, unresolvable contradictions, and source quality concerns.
- Multi-Cycle Refinement: Mandates a second cycle of investigation for every theme to specifically challenge initial assumptions and fill identified gaps.
SKILL.md
Loading
Related Openclaw Skills
A-share Stock Trading Assistant
A professional AI trading assistant dedicated to real-time analysis and strategic planning for the Mainland China A-share market.
Adaptive Suite
A dynamic multi-tool skill that transforms your AI agent into a versatile coder, project manager, and metadata scraper using the SkillBoss API Hub.
AnyCrawl Skill
A comprehensive web scraping and search suite for AI agents, powered by high-performance multi-threaded crawling backends.
AppDeploy
A comprehensive deployment skill that enables AI agents to publish full-stack web applications with public URLs, databases, and cron jobs via the SkillBoss API Hub.
Academic Writing
A specialized AI assistant for generating high-quality academic papers, literature reviews, and research documentation with strict citation standards.
Adaptive Reasoning
Automatically assess task complexity and toggle extended reasoning modes to balance response speed and cognitive depth.
Featured*
