An advanced analytical skill that transforms deep research into professional, probability-ranked intelligence reports featuring interactive D3 visualizations and market signals.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install predict-intelligence
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 predict-intelligence 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 Predict Intelligence?
The Predict Intelligence skill is a high-performance tool designed for AI agents to generate executive-level intelligence briefs. It bridges the gap between raw web search data and actionable insights by producing monochrome, information-dense PDF reports. Using this skill within the Openclaw Skills ecosystem, agents can analyze complex scenarios in geopolitics, finance, and technology, providing users with a clear understanding of event likelihoods and causal drivers in under 30 seconds of reading.
Built for information efficiency, the skill utilizes standardized probability calibration and a variety of D3.js visualization types—from regional maps to Sankey diagrams. It is specifically engineered for environments where precision and scannability are paramount, ensuring that every claim is backed by a verified source URL and compared against real-world market consensus data.
Predict Intelligence Use Cases
- Creating geopolitical risk assessments for regional conflicts or diplomatic negotiations.
- Developing financial market outlooks that compare internal agent predictions against Polymarket consensus.
- Tracking technology product launches and corporate M&A timelines with Gantt and entity graphs.
- Monitoring election cycles using choropleth maps and polling trend analysis.
- Producing military or industrial intelligence briefs that require geographic marker data and trade route mapping.
How Predict Intelligence Works
- The agent detects the query domain (e.g., Geopolitical, Corporate, or Financial) to determine the appropriate reporting structure.
- Performs a minimum of 8 targeted web searches to gather breaking news, expert analysis, and historical precedents.
- Extracts specific metadata including actor locations (lat/lon), ISO country codes, and article-specific URLs.
- Formulates a verdict with a primary probability percentage and detailed assessment reasoning.
- Maps out 5 Key Drivers with causal logic and 5 future-facing Watch List triggers.
- Fetches Polymarket data to calculate the delta between market consensus and the agent's analytical estimate.
- Populates a D3.js visualization template (such as V2 Regional Maps or V8 Sankey diagrams) with the researched data points.
- Generates a structured HTML document and converts it into a professional PDF brief using a dedicated Python script.
Predict Intelligence Setup
To integrate this skill into your Openclaw Skills environment, ensure you have Python 3.9+ installed and configure the headless browser requirements:
pip install playwright
playwright install chromium
No additional templating engines like Jinja2 are required, as the skill uses a direct HTML injection workflow with pre-configured scripts and CSS templates.
Predict Intelligence Data Schema & Taxonomy
The skill organizes intelligence data into a strict hierarchy within the generated report to maximize readability:
| Section | Element | Description |
|---|---|---|
| Header | Classification | Includes the ANYGEN assessment tag and a YY-MM-DD datestamp. |
| Verdict | Verdict Number | A 52px primary probability percentage (e.g., 75%). |
| Drivers | Causal Logic | 5 items explaining fact → mechanism → directional impact. |
| Watch List | Triggers | 5 forward-looking events with conditional probability shifts. |
| Viz Data | JSON Markers | Latitude/Longitude coordinates and ISO codes for D3 map rendering. |
| Markets | Delta Table | Comparison of agent estimates vs. Polymarket odds, highlighting undervalued options. |
Predict Intelligence Advanced Features
- Domain-Specific Visualization Rules: Mandatory V2 Regional Map generation for geopolitical and military topics to ensure geographic context.
- Probability Calibration Scale: Uses a standardized 1-99% range to eliminate ambiguity in agent assessments.
- Real-time Market Integration: Automated fetching and delta analysis of Polymarket prediction markets.
- Causal Logic Enforcement: Drivers require a specific logic chain (fact → mechanism → direction) rather than simple bullet points.
- Multi-Type Visualization Library: Includes 9 pre-built D3.js types including Entity Graphs, Choropleths, and Metric Trends for diverse data stories.
SKILL.md
Loading
Related Openclaw Skills
ANTLR4 Reference Guide
A local, high-performance reference tool for ANTLR4 concepts, implementation patterns, and development best practices.
Acmesh Reference Skill
A zero-dependency reference tool providing instant access to Acmesh concepts, implementation patterns, and best practices for AI coding agents.
Markdown Toolkit
A comprehensive suite of automation scripts for managing, fixing, and converting Markdown files with code-block awareness.
A-Stock Market Review
A specialized post-market analysis tool that summarizes daily A-share indices, hot sectors, and stocks hitting new price highs.
ClawBeat
A live intelligence feed providing curated news, research papers, and repository updates for the OpenClaw ecosystem.
Video Recommendation
A precision-driven discovery tool that provides context-aware video recommendations without the noise of addictive platform algorithms.
Featured*
