deep-researcher

What This Skill Does

The deep-researcher skill functions as a high-level orchestration agent designed for complex, evidence-based inquiry. Unlike basic web-search tools, this meta-skill decomposes broad queries into iterative sub-questions, systematically gathers data from multiple specialized domains, and synthesizes the findings into a rigorous, scientific-style Markdown report. It integrates deepresearchwork for logical structure, tavily-search for web-based context, perplexity-deep-search for reasoning-heavy analysis, and literature-search for academic verification via Semantic Scholar. By enforcing a hypothesis-driven approach, it ensures that your research not only answers the surface question but also accounts for potential contradictions, validates source quality, and provides clear, footnoted citations.

Installation

To install and configure the skill, ensure you have your API keys ready and run the following in your OpenClaw environment:

npx -y clawhub@latest install deepresearchwork tavily-search literature-search perplexity-deep-search
npx -y clawhub@latest update --all

Before execution, verify your environment credentials:

echo "$TAVILY_API_KEY"
echo "$PERPLEXITY_API_KEY"

Use Cases

This skill is ideal for:

• Writing comprehensive white papers on emerging technologies (e.g., fusion energy).
• Conducting literature reviews where academic accuracy is paramount.
• Resolving conflicting reports in market trends or geopolitical analysis.
• Investigating multi-faceted problems that require both real-time news and peer-reviewed studies.

Example Prompts

1. "Perform a deep research report on the state of carbon sequestration technology for the 2030 horizon, focusing on European policy impacts. Ensure at least 5 academic sources and a section on economic feasibility."
2. "I need a comparative analysis of LLM reasoning benchmarks vs. human cognitive models. Use the /semantic-scholar workflow for academic foundations and contrast this with recent industry white papers."
3. "Investigate the impact of sea-level rise on global supply chain logistics. Focus on the last 5 years, prioritize peer-reviewed findings, and flag any major discrepancies between regional government reports."

Tips & Limitations

• Strict Scoping: This skill requires explicit input parameters. Do not attempt to run it without defining your research_topic, target_horizon, and evidence_threshold.
• Mapping Note: If you specifically request a semantic-scholar search, the agent will automatically pivot to literature-search, as this is the designated provider for academic data within the OpenClaw ecosystem.
• Accuracy: While the agent handles contradiction resolution, always review the generated report for contextual nuance, especially on sensitive or highly subjective topics. The deepresearchwork component acts as a procedural guide; ensure you verify the underlying citations in the final output.