工具|提示词选择器·全部
A 503-entry dropdown of ready-made prompts, wired straight into your generator
- 提示词文本
- 卡片标题
- 卡片详情
Some days you don't want to write a prompt - you want to choose one. The Prompts Selector is the general-purpose picker for the Tikpan prompt library: a single dropdown holding every card across all nine synced repos (500+ entries once you've synced, the brief shows 503 choices), with repository and tag filters on the side. Pick a card, and out the other end comes the prompt text, the card title, and the full card details. It's the "全部" (all) variant - the pack also ships image-only and video-only selectors that do exactly this but pre-filtered.
This node is a convenience layer, not a generator. It doesn't call an API and needs no key. All it does is read the locally-synced card index and serve you a card. The value is in the curation: the library is built from repos that collect working, battle-tested prompts for specific models and use-cases (Nano Banana, GPT-Image-2, Seedance, video workflows), which is a genuinely faster start than a blank prompt box.
How it works
On the surface it's a giant enum widget. Underneath, the node reads the same data/prompts-library/ JSON index that the manager node syncs and the search node queries, and builds dropdown labels like 12. 头像 - 秋日环境人像 - 使用参考图像中完全相同的面部,生成超写实.... 过滤_仓库 and 过滤_标签 restrict which cards appear in the dropdown, and 搜索关键词 filters by keyword - so even a 500-item list becomes findable without switching to the dedicated search node. 显示详情 (default on) controls whether the full card text comes through in the detail output.
Inputs and outputs that matter
The 选择提示词 dropdown is the whole game. The three STRING outputs are 提示词文本 (wire this into any Tikpan image node's prompt input, or a GPT-Image-2/Nano Banana node), 卡片标题, and 卡片详情 (the full card body - useful if the card includes arguments you need to fill in, like {argument name="subject"}). The pack's own quick-start shows the pattern: [提示词选择器] → 提示词文本 → [生图节点].
Installing
Pack install as usual - Manager search "Tikpan", or git clone https://github.com/htrert/ComfyUI-Tikpan-Pro into custom_nodes, restart. Then the one prerequisite that trips everyone: run the Prompts Manager (同步全部仓库) first. Empty library, empty dropdown - the node even shows a "(请先同步提示词库)" hint when there's nothing to show.
Common issues
- Dropdown is empty or says "请先同步". You haven't synced. Run the manager node once. This is the #1 support moment for the whole library feature.
- The card you picked doesn't match the prompt you got. Some cards contain template arguments like
{subject}- the library treats those literally. Fill them in before sending to the generator, or the model gets a literal{subject}token. - Filters hide everything. 过滤_仓库 + 过滤_标签 + 搜索关键词 stack as ANDs. If your combo returns nothing, loosen one of them.
- Library staleness. The dropdown only knows what's synced. Re-run the manager periodically if you rely on fresh repos - curated prompts are a snapshot, and prompt conventions drift with every model release (prompt-engineering.md).
Inputs (5)
| Name | Type | Default | Description |
|---|---|---|---|
| 选择提示词 | COMBO | 1. No. 1: 头像/虚拟形象 - 电影感柔和粉彩棚拍肖像 — 高分辨率、柔和电影感的年轻女性肖像,她倚靠在一片{ | 503 options: 1. No. 1: 头像/虚拟形象 - 电影感柔和粉彩棚拍肖像 — 高分辨率、柔和电影感的年轻女性肖像,她倚靠在一片{, 2. No. 2: 头像/虚拟形象 - 现代建筑时尚肖像 — 中等半身肖像,一位肤色为橄榄色、晒成古铜色的肌肉成年男性,, 3. No. 3: 头像/虚拟形象 - 撕裂社交媒体头像艺术 — 一幅极具创意的数字艺术作品,其中{argument name="subject", 4. No. 4: 头像/虚拟形象 - 奢侈时尚杂志封面肖像 — 使用我上传的面部图像作为主要身份参考。保留我精确的, 5. No. 5: 头像/虚拟形象 - 优雅酒店橱窗时尚大片 — 创作一幅受高端亚洲杂志启发的超现实主义奢华时尚肖像,, 6. No. 6: 头像/虚拟形象 - 高保真身份肖像模板 — 使用上传的面部图像作为主要身份参考。保留精确的, +497 |
| 过滤_仓库opt | COMBO | 全部 | 10 options: 全部, YouMind-OpenLab/awesome-nano-banana-pro-prompts, EvoLinkAI/awesome-gpt-image-2-API-and-Prompts, EvoLinkAI/awesome-seedance-2.0-prompts, EvoLinkAI/awesome-seedance-2-guide, toki-plus/ai-mixed-cut, +4 |
| 过滤_标签opt | COMBO | 全部 | 9 options: 全部, director, gpt-image-2, guide, image, nano-banana, +3 |
| 搜索关键词opt | STRING | — | |
| 显示详情opt | BOOLEAN | true | — |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| 提示词文本 | STRING | — |
| 卡片标题 | STRING | — |
| 卡片详情 | STRING | — |