A conversational AI trading agent that uses Fuku sports models to scan and execute trades on Kalshi prediction markets based on natural language preferences.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install fuku-predictions
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 fuku-predictions 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 fuku-predictions?
Fuku Predictions is a sophisticated trading integration designed for Openclaw Skills that allows users to engage with Kalshi prediction markets through a conversational interface. It leverages professional sports models for NBA, CBB, NHL, and Soccer to identify market inefficiencies and calculate statistical edges. By synthesizing complex data into actionable insights, this skill empowers traders to build personalized profiles that reflect their specific risk tolerance and situational preferences.
The skill prioritizes security and local control, ensuring that sensitive Kalshi API keys never leave the user's environment. Whether you are looking for a manual market browser or a fully autonomous trading bot, Fuku Predictions provides the tools necessary to navigate high-frequency prediction markets with precision and institutional-grade edge analysis.
fuku-predictions Use Cases
- Creating custom trading profiles using natural language like "Find home dogs getting 7+ points."
- Automated scanning of sports markets to find the highest statistical edge for NBA and CBB games.
- Hands-free autonomous trading within strict risk limits and daily loss caps.
- Monitoring Kalshi positions and portfolio balance through a simplified command-line or chat interface.
- Identifying player mismatches and statistical outliers across multiple professional sports leagues.
How fuku-predictions Works
- The agent initializes by loading local environment variables and RSA private keys for secure Kalshi API authentication.
- Users define their trading criteria using natural language, which the profile engine converts into a structured JSON configuration.
- The skill fetches real-time sports predictions from the public Fuku API and compares them against current Kalshi market prices.
- It calculates the edge using normal distribution probability conversion and sport-specific standard deviations.
- Recommended trades are presented in three tiers (Safer, Main, and Riskier) with clear dollar-to-contract math.
- Upon user approval or autopilot triggers, the skill signs and transmits orders to Kalshi using RSA-PSS signatures.
fuku-predictions Setup
First, install the required Python dependencies:
pip install httpx cryptography python-dotenv
Create a .env file in the skill directory with your credentials from the Kalshi API profile page:
KALSHI_API_KEY_ID=your_key_id
KALSHI_PRIVATE_KEY="-----BEGIN RSA PRIVATE KEY-----\n...\n-----END RSA PRIVATE KEY-----"
Run the setup wizard to verify your connection:
python3 scripts/setup.py
fuku-predictions Data Schema & Taxonomy
The skill maintains a structured local data environment to manage trades and preferences safely:
| Directory/File | Data Type | Description |
|---|---|---|
config/profiles/*.json |
User Profiles | Stores natural language-derived trading logic and filters. |
config/config.json |
Strategy Config | Defines global risk limits, sizing (e.g., Quarter-Kelly), and operation modes. |
trades.json |
Transaction Log | A local record of all order executions, payouts, and historical performance. |
KILL_SWITCH |
Safety Trigger | An empty file that, if present, immediately halts all autonomous trading activity. |
fuku-predictions Advanced Features
- Natural Language Profile Building: Translates complex situational betting strategies into executable filters.
- Multi-Sport Edge Math: Utilizes specific sigma values for spreads and totals across NBA, CBB, NHL, and Soccer.
- Autopilot Modes: Supports
dry_runfor testing,approvefor interactive trading, andautofor full automation. - RSA-PSS Security: Local signing of API requests ensures that private keys are never transmitted over the network.
- Three-Tier Market Display: Automatically identifies the safest high-edge contracts versus high-payout risky contracts for any given game.
SKILL.md
Loading
Related Openclaw Skills
Wikclawpedia
A programmatic gateway to the living archive of the AI agent renaissance, enabling agents to search, retrieve, and submit historical intel.
Apple Calendar Manager
A specialized macOS skill for programmatically managing Apple Calendar events using AppleScript and smart date parsing.
nova-letters
A reflective journaling tool for AI agents to maintain context and capture meaningful insights across sessions.
Drift
Drift is a specialized conversation protocol that allows AI agents to maintain continuity by leaving questions and reflections for future instances of themselves.
Fuku Sportsbook Agent Skill
An automated sports betting and data analysis skill for creating AI agents that track odds, post picks, and manage virtual or real USDC bankrolls.
Game Developer
A specialized AI agent skill designed to architect, implement, and optimize high-performance games using Unity, Unreal Engine, and ECS patterns.
Featured*
