synapse-plugin-execution

Execute Synapse plugin actions via run_plugin() in local, Ray Actor, or Ray Job modes.

1|1|Updated Jan 14, 2026
One-click install
npx skills add https://github.com/datamaker-kr/synapse-claude-marketplace --skill synapse-plugin-execution
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: synapse-plugin-execution
Source: https://github.com/datamaker-kr/synapse-claude-marketplace/tree/main/plugins/synapse-plugin-helper/skills/plugin-execution
Command: npx skills add https://github.com/datamaker-kr/synapse-claude-marketplace --skill synapse-plugin-execution

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill simplifies the process of running Synapse plugins and their actions directly from your code, enabling seamless integration into custom workflows and automated processes.

Core Features & Use Cases

  • Programmatic Execution: Run plugin actions using run_plugin() with various execution modes (Local, Task, Job).
  • Plugin Discovery: Inspect and load plugins from filesystem paths or Python modules using PluginDiscovery.
  • Use Case: Integrate a Synapse plugin's training action into a larger MLOps pipeline, triggering it with specific parameters and execution modes based on pipeline stage.

Quick Start

Use the plugin execution skill to run the 'train' action from the '/path/to/plugin' directory with 'epochs' set to 100 in local mode.

Frequently Asked Questions about synapse-plugin-execution

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute Synapse plugins programmatically from my Python code?

You can programmatically execute Synapse plugins by calling the run_plugin() function, which provides a unified interface to trigger specific plugin actions with custom parameters directly from your automation scripts.

What execution modes are available for running Synapse SDK plugin actions?

Synapse SDK plugin actions support three execution modes for flexible deployment: Local mode for direct execution, Ray Actor mode for distributed tasks, and Ray Job mode for submitting separate jobs.

How does plugin discovery work when integrating Synapse plugins dynamically?

Plugin discovery uses the PluginDiscovery component to inspect and load available Synapse plugins from filesystem paths or Python modules, enabling dynamic integration into custom applications without hardcoded imports.

Can I trigger a Synapse plugin training action as part of an MLOps pipeline?

Yes, you can trigger a Synapse plugin training action within an MLOps pipeline by using run_plugin() to pass specific parameters like epochs and select an appropriate execution mode based on your pipeline stage.

Do I need Ray to run Synapse plugins in local mode?

No, you do not need Ray to run Synapse plugins in local mode. The local execution mode allows you to run plugin actions directly without requiring a distributed Ray cluster setup.