Ganglion Synth City - SN50 Mining for Openclaw

A specialized Openclaw Skills package for mining Bittensor Subnet 50 through probabilistic price-path forecasting and CRPS scoring.

tensorlink-dev
v0.1.2
Mar 5, 2026
0
889
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install skill-8

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install skill-8 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 Ganglion Synth City - SN50 Mining?

Ganglion Synth City is a technical domain skill designed for miners on the Bittensor Synth subnet (netuid 50). It provides the necessary logic to generate 1,000 Monte Carlo price paths per asset, enabling miners to participate in both low-frequency (24h) and high-frequency (1h) forecasting competitions. By integrating this skill, users can leverage sophisticated financial modeling techniques to predict price movements for assets like BTC, ETH, and even equity-based synthetic assets.

Built to work seamlessly with Openclaw Skills, this tool handles the complexities of the Continuous Ranked Probability Score (CRPS) pipeline, which validators use to rank submissions. It bridges the gap between raw data from Pyth Hermes and the specific submission formats required by the Bittensor protocol, ensuring miners can focus on refining their volatility estimation and simulation strategies rather than infrastructure overhead.

Ganglion Synth City - SN50 Mining Use Cases

  • Generating probabilistic price forecasts for the Bittensor SN50 mining competition.
  • Estimating asset volatility using EWMA or GARCH models to improve scoring accuracy.
  • Backtesting forecasting strategies against historical Pyth Benchmarks data before on-chain submission.
  • Automating multi-asset path generation for high-impact assets like XAU and SPYX.
  • Calibrating Monte Carlo simulations to match specific validator scoring intervals and horizons.

How Ganglion Synth City - SN50 Mining Works

  1. The miner initializes the environment using the Openclaw Skills framework for SN50.
  2. Live spot prices are fetched via the Pyth Hermes oracle to set the simulation baseline.
  3. Historical price data is retrieved to estimate sigma values and identify recent volatility regimes.
  4. The skill executes Monte Carlo simulations (e.g., Geometric Brownian Motion) to generate 1,000 distinct price paths.
  5. Paths are formatted into a precise tuple structure including the start timestamp and time increment.
  6. The resulting ensemble is submitted to the subnet validator, where it is scored using the CRPS pipeline across multiple intervals.

Ganglion Synth City - SN50 Mining Setup

To deploy this skill within your Openclaw Skills environment, ensure you have ganglion and curl installed. Follow these steps to initialize and run the miner:

# Initialize the project scaffold
ganglion init ./synth-city --subnet sn50 --netuid 50

# Set the project environment and check status
export GANGLION_PROJECT=./examples/synth-city
ganglion status $GANGLION_PROJECT

# Execute the mining skill
ganglion run $GANGLION_PROJECT

Ganglion Synth City - SN50 Mining Data Schema & Taxonomy

The skill manages data across simulations and historical lookups to align with the validator's scoring requirements. Below is the primary data structure for submissions:

Parameter Description Constraints
start_timestamp Unix seconds of the request Must match validator input
time_increment Interval in seconds (60 or 300) Must match validator input
price_paths 1,000 arrays of price points Max 8 significant digits per value
asset_coeffs Weighting multipliers per asset XAU=2.262, SPYX=2.991, BTC=1.000

Ganglion Synth City - SN50 Mining Advanced Features

  • Support for Merton Jump-Diffusion and Regime-Switching models to capture non-normal price action.
  • Full validator scoring pipeline replay for precise local backtesting and CRPS optimization.
  • Automated handling of NaN gaps in historical data using the label_observed_blocks logic.
  • Ensemble blending capabilities to combine paths from multiple statistical models.
  • Real-time volatility estimation from recent returns to replace static sigma baselines.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Github Stars: 0
forks: 0

Featured*