jtydhr88/comfyui-custom-node-skills

A curated collection of [Claude Code skills](https://docs.anthropic.com/en/docs/claude-code/skills) for developing ComfyUI custom nodes. These skills give Claude comprehensive knowledge of the ComfyUI node system, covering both the V3 (recommended) and V1 (legacy) APIs.

What it solves

Developing custom nodes for ComfyUI can be complex due to the specific requirements of its API, node structure, and frontend integration. This project provides a set of standardized "skills" that teach AI coding agents how to write, migrate, and package ComfyUI nodes correctly, ensuring they follow best practices for both the modern V3 API and legacy V1 systems.

How it works

The project consists of a collection of SKILL.md files based on the open agentskills.io standard. When installed into an AI agent (like Claude Code, OpenAI Codex, or GitHub Copilot), these files provide the agent with deep, verified knowledge of the ComfyUI codebase. The agent can then use this context to assist you with tasks like defining node inputs, handling data types, creating frontend UI extensions, or migrating older nodes to the current V3 standard.

Who it’s for

This is for developers building custom nodes for ComfyUI who want to use AI coding assistants to speed up their workflow. It is ideal for those who want to ensure their code is accurate, follows the latest API patterns, and adheres to official ComfyUI implementation standards.

Highlights

  • V3 API Focused: Prioritizes the modern V3 API while providing legacy V1 support for migration.
  • Multi-Agent Support: Works across Claude Code, OpenAI Codex, and GitHub Copilot using a single, standardized format.
  • Source-Verified: All skills are cross-referenced against the actual ComfyUI backend and frontend source code.
  • Comprehensive Coverage: Includes everything from basic node setup and data types to advanced dynamic inputs and frontend JavaScript extensions.
  • Easy Integration: Offers simple installation methods via plugin marketplaces or direct file copying.

Related

  • Project
    PlamenTSV/plamen

    An autonomous AI security auditor for smart contracts and L1 node-client infrastructure that orchestrates AI agents to produce verified audit reports with PoC exploits.

  • Project
    microsoft/debug-gym

    A text-based interactive debugging framework for Python programs that allows LLM agents to use tools like pdb and bash to investigate and fix bugs in code repositories.

  • Project
    robzilla1738/harness-terminal

    A native macOS terminal with GPU acceleration and session persistence that integrates with AI coding agents to notify users when they need attention.

  • Project
    AlgoNoRhythm/Flare

    A graph-first IDE for agentic coding that provides a live visual map of a codebase and a coordination layer for multiple AI agents to work together without colliding.