A specialized skill for comparing, contrasting, and synthesizing findings across multiple completed DeepScan research sessions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install comparative-synthesis
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 comparative-synthesis 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 Comparative Synthesis?
Comparative Synthesis is a technical extension for AI coding agents within the Openclaw Skills ecosystem. It is designed to move beyond single-session monitoring by pulling historical data from various DeepScan runs to identify overlapping papers, conflicting findings, and complementary themes. This skill empowers users to extract high-level intelligence from their entire research history rather than viewing sessions in isolation.
By leveraging existing report data, this skill facilitates deep meta-analysis without the need for redundant web searches or new scans. It bridges the gap between raw data collection and strategic insight, making it an essential component for researchers and developers who require a unified view of complex literature landscapes.
Comparative Synthesis Use Cases
- Comparing findings between two specific DeepScan runs on related topics.
- Synthesizing themes and recurring patterns across a month's worth of research sessions.
- Generating side-by-side visualizations of citation counts or publication years from different reports.
- Identifying knowledge gaps where previous research sessions failed to cover specific topics.
How Comparative Synthesis Works
- The agent calls the summarize_evidence tool to aggregate high-level summaries across the user's stored history.
- For specific comparisons, the agent retrieves full report data using the get_deepscan_report tool.
- The system identifies divergences and commonalities in papers, methods, and conclusions.
- When quantitative data is available, the agent uses run_python_plot to generate visual comparisons such as Venn diagrams or histograms.
- A final synthesis report is generated, structured around common ground, divergences, and emerging trends.
Comparative Synthesis Setup
To integrate this skill into your agent environment, ensure the core DeepScan tools are configured within your Openclaw Skills directory.
# Install the Openclaw Skills repository
git clone https://github.com/openclaw/skills.git
# Navigate to the skill directory
cd skills/comparative-synthesis
# Ensure environment variables for report storage are set
export DEEPSCAN_HISTORY_PATH="/path/to/your/reports"
Comparative Synthesis Data Schema & Taxonomy
The skill organizes research data through a structured analysis of existing JSON and Markdown reports. The synthesis process follows this taxonomy:
| Feature | Description |
|---|---|
| Common Ground | Unified list of papers and findings present across multiple runs. |
| Divergences | Highlights where different sessions reached conflicting conclusions. |
| Gaps | Identifies requested topics that are missing from all processed reports. |
| Trends | Temporal patterns and shifting consensus identified via metadata analysis. |
Comparative Synthesis Advanced Features
- Visual Meta-Analysis: Automated generation of topic co-occurrence heatmaps and citation distributions using Python plotting.
- Aggressive Condensation: Uses summarize_evidence for rapid scanning of large research volumes.
- Temporal Pattern Recognition: Tracks how emerging methods or shifting consensus evolve across different timestamps in the user's research history.
- Multi-Report Cross-Referencing: Simultaneously handles full-text data from multiple specified report IDs for granular side-by-side analysis.
SKILL.md
Loading
Related Openclaw Skills
LinkdAPI LinkedIn Integration
A comprehensive LinkedIn data integration for Openclaw Skills providing access to over 50 endpoints for professional networking and B2B data.
Ping-Me Reminder Skill
A natural language reminder system that enables one-shot notifications across multiple chat channels with automatic timezone and channel detection.
Deep Night Treehole
Deep Night Treehole is an empathetic AI companion within the Openclaw Skills ecosystem designed to provide a safe, non-judgmental space for emotional expression and validation.
Sentiment Bot
A sophisticated AI agent for detecting sentiment, intensity, and emotional nuances across text, URLs, and batch datasets.
Product Analytics
A comprehensive framework for defining, tracking, and interpreting product metrics across every stage of the product lifecycle.
PricingCalc Product & Service Pricing Calculator
A professional financial utility for calculating accurate service rates and product prices based on costs, competition, and target margins.
Featured*
