Extensions/ComfyUI-PromptPresetManager
ComfyUI Extension

ComfyUI-PromptPresetManager

A ComfyUI extension with 1 custom node.

By yurishk·Created 16 days ago·Updated 10 days ago· 2
yurishk/ComfyUI-PromptPresetManager
Nodes1
On cloudLocal install
Categorytext/presets
Stars2
Updated10 days ago
Readme

Prompt Preset Manager

English | 简体中文

A shared, searchable prompt preset library for ComfyUI. Keep reusable prompts organized without turning each workflow into a wall of text nodes.

Prompt Preset Manager quick editor

Features

  • Shared preset library: presets are stored locally and are available across workflows.
  • Fast in-node editing: search, load, edit, favorite, overwrite, or save a variant without leaving the node.
  • Full library manager: organize presets with nested folders, types, tags, descriptions, favorites, filters, and sorting.
  • Workflow-safe drafts: the selected preset and local edits are serialized with the node, including duplicated nodes and reopened workflows.
  • Native text output: outputs one standard STRING, so it works with normal ComfyUI prompt inputs.
  • Upstream text capture: connect any STRING output to the optional text input; non-empty results are written back into the editable prompt box after execution.
  • Import and export: back up or move the whole library, a folder, a filtered view, or an individual preset.
  • English and Chinese UI: follows the ComfyUI/browser locale automatically.

Prompt Preset Manager node

Preset library manager

Installation

ComfyUI Manager

Search for Prompt Preset Manager and install it, then restart ComfyUI.

Manual

cd ComfyUI/custom_nodes
git clone https://github.com/yurishk/ComfyUI-PromptPresetManager.git

Restart ComfyUI after cloning. The plugin has no third-party Python dependencies.

Usage

  1. Add Prompt Preset Manager from text/presets.
  2. Open Preset Library, create folders and presets, or import an existing library.
  3. Search and select a preset in the node.
  4. Edit the native prompt box when a workflow needs a local variation.
  5. Optionally connect an upstream STRING to the text input to capture and continue editing its result.
  6. Connect the text output to any node that accepts a STRING prompt, or leave it unconnected and run this node as a text capture endpoint.

The shared library is stored in data/presets.json. Automatic backups are written to backups/. Both directories are local user data and are excluded from Git.

License

MIT


简体中文

English | 简体中文

一个适用于 ComfyUI 的全局提示词预设管理器。它可以集中管理常用提示词,避免每个工作流都堆满重复的文本节点。

预设管理器快速编辑区

功能

  • 全局预设库:预设保存在本地文件中,可在不同工作流之间共享。
  • 节点内快速编辑:无需离开节点即可搜索、载入、编辑、收藏、覆盖或另存变体。
  • 完整管理界面:支持多级文件夹、类型、标签、说明、收藏、筛选与排序。
  • 工作流持久化:当前选择和本地修改会随节点保存,复制节点、切换或重新打开工作流后仍可恢复。
  • 原生文本输出:仅输出标准 STRING,可直接连接 ComfyUI 的普通提示词输入。
  • 上游文本接收:可将任意 STRING 输出连接到可选的 text 输入;执行后的非空结果会写回原生提示词框,方便继续修改。
  • 导入与导出:可备份或迁移整个预设库、文件夹、筛选结果或单个预设。
  • 中英文界面:根据 ComfyUI 或浏览器语言自动切换。

预设管理器节点

完整预设库管理界面

安装

ComfyUI Manager

搜索 Prompt Preset Manager 并安装,然后重启 ComfyUI。

手动安装

cd ComfyUI/custom_nodes
git clone https://github.com/yurishk/ComfyUI-PromptPresetManager.git

克隆完成后重启 ComfyUI。本插件不需要安装额外的 Python 依赖。

使用方法

  1. text/presets 分类添加 Prompt Preset Manager
  2. 打开“预设库”,创建文件夹和预设,或导入已有预设库。
  3. 在节点中搜索并选择预设。
  4. 某个工作流需要单独调整时,直接编辑节点内的原生提示词文本框。
  5. 如需接收提示词强化等上游结果,将其 STRING 输出连接到本节点的 text 输入。
  6. 可将本节点的 text 输出继续连接到下游,也可以不连接,将本节点直接作为文本接收终点运行。

全局预设库保存在 data/presets.json,自动备份位于 backups/。这两个目录属于本地用户数据,不会提交到 Git。

许可证

MIT