Ganglion: Bittensor Mining Execution Engine for Openclaw

Ganglion is a domain-specific execution engine designed to orchestrate autonomous mining agents within Bittensor subnets.

tensorlink-dev
v1.0.0
Mar 4, 2026
0
1.1k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install codebase-guide

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 codebase-guide 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: Bittensor Mining Execution Engine?

Ganglion serves as a robust search infrastructure for Bittensor subnet mining, providing a sophisticated pipeline framework to manage autonomous agents. Instead of defining the models themselves, it focuses on the search process, coordinating how agents discover optimal configurations. As a key component in the ecosystem of Openclaw Skills, it offers a flexible architecture through its CLI, Python library, and HTTP bridge API.

The framework is built for durability and growth, featuring a cross-run strategic memory that compounds over time. By allowing agents to access shared knowledge of successful patterns and failed anti-patterns, it enables a level of emergent cooperation essential for competitive mining environments.

Ganglion: Bittensor Mining Execution Engine Use Cases

  • Orchestrating autonomous mining agents for Bittensor subnet model search.
  • Scaling mining operations across multiple bots using a shared knowledge pool.
  • Automating complex multi-stage pipelines with built-in retry and escalation policies.
  • Integrating external tools via MCP to expand agent capabilities at runtime.
  • Auditing and rolling back configuration changes in remote mining environments.

How Ganglion: Bittensor Mining Execution Engine Works

  1. The user initializes a project directory containing a config.py that defines the subnet configuration and pipeline structure.
  2. The engine detects the operating mode (Local or Remote) based on environment variables like GANGLION_URL or GANGLION_PROJECT.
  3. The orchestrator executes defined pipeline stages in dependency order, utilizing specific agents and registered tools.
  4. Strategic memory is captured in a cross-run knowledge store, allowing agents to learn from successful patterns and previous failures.
  5. The system continuously applies retry policies and manages state mutations, which can be audited or rolled back via the API for high reliability.

Ganglion: Bittensor Mining Execution Engine Setup

Ensure you have Python 3.11+ and your OPENAI_API_KEY configured. Install the package and initialize your project to begin using Openclaw Skills:

pip install ganglion
ganglion init ./my-subnet --subnet sn9 --netuid 9

To run in remote mode, start the HTTP bridge:

ganglion serve ./my-subnet --bot-id alpha --port 8899
export GANGLION_URL=http://127.0.0.1:8899

Ganglion: Bittensor Mining Execution Engine Data Schema & Taxonomy

Ganglion organizes data within a project directory structure. The primary configuration resides in config.py, while operational state is tracked through an audit log and knowledge store.

Component Description
config.py Defines SubnetConfig and PipelineDef for the engine.
tools/ Directory containing local tool implementations.
agents/ Directory for agent logic and prompt templates.
knowledge/ Cross-run memory store indexed by capability and bot ID.
audit log Records all mutations for validation and multi-step rollback.

Ganglion: Bittensor Mining Execution Engine Advanced Features

  • Multi-Bot Workflows: Share strategic knowledge across multiple Openclaw Skills sessions using unique bot identifiers for cooperative mining.
  • MCP Integration: Connect to external Model Context Protocol servers to dynamically inject new tools into the agent repertoire at runtime.
  • Runtime Mutation: Patch pipelines, register agents, or swap retry policies via API without stopping the execution engine.
  • Model Escalation: Automatically switch to more powerful LLMs or climb a cost ladder when cheaper models fail to make progress.
  • Standardized Response Envelope: High-level error handling and data extraction using a consistent success/error JSON structure.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins python3ganglion
Github Stars: 0
forks: 0

Featured*