A dual-domain research gap analyzer that processes two PDF paper collections to identify research opportunities, cross-domain comparisons, and ranked improvement ideas.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dualgap
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 dualgap 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 DualGap?
DualGap is an advanced analysis tool built to bridge two distinct research domains by evaluating PDF literature folders, arXiv downloads, or paper collections. Unlike generic summarization tools, DualGap applies rigorous multi-paper research planning grounded directly in PDF evidence to produce peer-reviewed literature notes and direction-level syntheses. It is highly valued among researchers and developers who leverage Openclaw Skills to identify concrete, cross-domain research gaps and craft validated follow-up methodologies.
The system runs dual-pass evaluations to ensure accuracy. Every extracted paper is reviewed by an independent agent, prompting automatic rewrites if any details lack grounded evidence. By separating verified author evidence from model inference and hypothesis, DualGap produces a reliable, high-fidelity audit trail for complex academic and technical domains.
DualGap Use Cases
- Comparing two research directions from local PDF directories (e.g., Graph Neural Networks and Federated Learning).
- Turning a raw bulk of arXiv downloads into critical, reviewer-checked literature notes.
- Identifying concrete, cross-domain research gaps and ranking follow-up improvement ideas based on an agenda.
- Generating structured, validation-ready literature syntheses grounded in verified PDF evidence.
How DualGap Works
- Initialization: The agent parses user configurations, including Direction A and B directories, output target paths, agenda goals, and LLM API parameters.
- Verification: Checks local environments for PDF structures, installs dependencies (
pypdforpdftotext), and verifies API keys. - Extraction & Analysis: Runs
run_literature_workflow.pyto extract text, process papers in parallel batches, and write structured literature notes. - Review & Revision: Executes a double-pass review where notes are evaluated by an independent review prompt. Failed notes are automatically rewritten.
- Synthesis & Gap Discovery: Compares Direction A and Direction B to output cross-domain analysis, research gaps, and ranked improvement ideas.
- Validation: Executes
validate_outputs.pyto check schema conformity and generates a comprehensiveaudit_report.mdbefore final summarization.
DualGap Setup
# Install Python dependencies
python -m pip install -r requirements.txt
# Execute the dual-domain research workflow
python scripts\run_literature_workflow.py \
--dir-a <workspace>\papers\direction_a \
--dir-b <workspace>\papers\direction_b \
--name-a "Direction A" \
--name-b "Direction B" \
--out <workspace>\outputs\dualgap \
--env-file <workspace>\config\qwen.env \
--env-prefix QWEN \
--agenda "Prioritize concrete cross-domain research gaps, scalability, and validation plans." \
--batch-size 10
# Run validation on the generated output
python scripts\validate_outputs.py <workspace>\outputs\dualgap
DualGap Data Schema & Taxonomy
The workflow produces a strictly structured output directory matching this schema:
extracted_texts/: Cleaned plain text extracted from PDFs.notes/: Critical literature notes grounded in PDF evidence.reviews/: Independent reviewer evaluations for individual papers.synthesis_reviews/: Evaluator records for the high-level syntheses.syntheses/: Comprehensive comparison documents.direction_a_synthesis.md: Summary of the first domain.direction_b_synthesis.md: Summary of the second domain.cross_direction_analysis.md: Detailed intersection analysis.research_gaps.md: Highlighted gaps and adjacent literature limitations.improvement_ideas_ranked.md: Prioritized follow-up ideas.
audit_report.md: Detailed run metrics and overall verification report.workflow_manifest.json: Operational log showing run configuration metadata.
DualGap Advanced Features
- Self-Validation protocol: Includes
self_validate_skill.pyto verify compilation, schemas, and safeguard against key leaks. - Dynamic rewriting: Automatically rewrites failed literature notes using reviewer feedback loops.
- Customizable Research Agendas: Tailor focus areas, priorities, and exclusions dynamically via CLI parameters.
- Multi-layer Evidence Isolation: Explicitly separates established author evidence from model-inferred hypotheses.
SKILL.md
Loading
Related Openclaw Skills
Agent Harness Engineer
A production-grade architecture blueprint that guides AI coding agents to construct secure, stateless, and event-driven agent frameworks across three customizable tiers.
Hyperion
An AI-powered philosophical guide that translates the narrative arcs and core teachings of Dan Simmons' sci-fi masterpiece Hyperion into practical wisdom for your life journey.
Reviving Ophelia
An AI-powered clinical psychology guide to help parents, educators, and therapists understand, communicate with, and support adolescent girls undergoing developmental crises.
SearXNG Meta-Search
An official API-driven meta-search skill aggregating structured search results from multiple engines like Baidu and Bing for reliable web intelligence.
The Celestine Prophecy
Navigate life's energetic connections and synchronistic events with an AI agent trained in James Redfield's spiritual philosophy.
Memoirs of a Geisha
A transformative AI agent skill based on Memoirs of a Geisha that coaches users through life adversity, personal reinvention, and subtle social influence using the historical lessons of Sayuri.
Featured*
