According to AlphaMoat, as of 2026-09, PPIO Multimodal Skill has 910 downloads and 1 stars — ranked #5,058 of 5,078 AI skills overall, and #510 of 512 in Design & Media.
| Platform | Installs | Downloads | Stars | Comments |
|---|---|---|---|---|
| ClawHub ↗ | 23 | 879 | 1 | - |
| SkillHub ↗ | 259 | 910 | 1 | — |
Platforms count differently and update at different times.
It covers text-to-image, image editing, text-to-video, image-to-video, TTS, and STT. Resolve the API key in priority order from the user message, ~/.ppio/config.json, or the PPIO_API_KEY environment variable; if none is available, provide setup guidance. Calls use the PPIO API with Bearer authentication and the corresponding Seedream, Vidu, MiniMax, or GLM service, with async task-result lookup. Check required inputs and parameters before submission and never claim a task completed without a result.
Profile published 2026-10-27
Generated from the skill’s documentation as examples; not results of actual runs.
The counting method changed in 2026-09: skills with the same name are now counted separately by author, and the full catalog is covered. Dashed lines show months before the change; they are not compared with later months, and no growth is computed across the change.
| Month | Downloads | MoM | Stars |
|---|---|---|---|
| 2026-09 | 910 | — | 1 |
| 2026-08 | 813 | +5.9% | 1 |
| 2026-07 | 768 | +23.7% | 1 |
| 2026-06 | 621 | +8.8% | 1 |
| 2026-05 | 571 | — | 1 |
| 2026-03 | 395 | — | 1 |
The author’s original file, shown as plain text without edits or review by AlphaMoat. Links are not clickable; check commands before running them. Fetched 2026-10-23.
--- name: ppio-multimodal description: | 使用 PPIO 执行多模态任务:文生图、图生图、文生视频、图生视频、TTS、STT。 适用于:生成图片、生成视频、文字转语音、语音识别。 --- # PPIO 多模态执行 ## 配置方式(三选一,按优先级)
### 方式1:配置文件(推荐)
创建文件 `~/.ppio/config.json`:
```json
{
"api_key": "你的API_Key"
}
```
**一条命令完成配置:**
```bash
mkdir -p ~/.ppio && echo '{"api_key": "你的API_Key"}' > ~/.ppio/config.json
```
### 方式2:环境变量
```bash
export PPIO_API_KEY="你的API_Key"
```
### 方式3:直接传参
在请求中直接提供:`请用 API Key sk_xxx 生成一张图片...`
---
## API Key 读取逻辑
```
1. 检查用户消息中是否包含 API Key(sk_ 开头)
2. 检查配置文件 ~/.ppio/config.json
3. 检查环境变量 PPIO_API_KEY
4. 都没有 → 返回配置引导
```
**配置引导消息(仅在未配置时显示):**
```
您还没有配置 PPIO 的 API Key。
快速配置(复制运行):
mkdir -p ~/.ppio && echo '{"api_key": "你的Key"}' > ~/.ppio/config.json
获取 Key:https://ppio.com/settings/key-management
```
---
## 执行流程(重要!)
```
用户请求 → 识别任务 → 获取 Key → ⚠️ 先发提示 → 执行任务 → 返回结果
```
### ⚠️ 必须先发送进度提示
**在调用 API 之前,必须先回复用户一条消息:**
```
🎨 收到!正在为您生成图片...
任务类型:文生图
使用模型:Seedream 5.0 Lite
预计耗时:5-15秒
预计费用:约 ¥0.245
请稍等,生成完成后会立即发送给您 ⏳
```
**这条消息必须在执行 API 调用之前发送!** 这样用户就知道任务已经开始处理,不会以为系统卡住了。
### 不同任务的提示模板
**文生图:**
```
🎨 收到!正在为您生成图片...
使用模型:Seedream 5.0 Lite
预计耗时:5-15秒
```
**文生视频:**
```
🎬 收到!正在为您生成视频...
使用模型:Vidu Q3 Pro
预计耗时:1-3分钟(视频生成较慢,请耐心等待)
```
**TTS:**
```
🔊 收到!正在为您生成语音...
使用模型:MiniMax Speech 2.8 Turbo
预计耗时:5-15秒
```
### 完成后的回复
```
✅ 生成完成!
[图片/视频/音频 URL]
实际消耗:¥0.245
```
### 视频任务的轮询提示
视频生成需要轮询,每 15 秒更新一次状态:
```
🎬 视频生成中...
当前状态:处理中
已等待:30 秒
预计还需:1-2 分钟
```
---
## API 配置
| 配置项 | 值 |
|--------|-----|
| Base URL | `https://api.ppio.com` |
| 鉴权 | `Authorization: Bearer <API_KEY>` |
| 获取 Key | https://ppio.com/settings/key-management |
## 任务类型与端点
| 任务 | 端点 | 模型 |
|------|------|------|
| 文生图 | `/v3/seedream-5.0-lite` | Seedream 5.0 Lite |
| 图片编辑 | `/v3/seedream-5.0-lite` | Seedream 5.0 Lite |
| 文生视频 | `/v3/async/vidu-q3-pro-t2v` | Vidu Q3 Pro |
| 图生视频 | `/v3/async/vidu-q3-pro-i2v` | Vidu Q3 Pro |
| TTS | `/v3/async/minimax-speech-2.8-turbo` | MiniMax Speech 2.8 |
| STT | `/v3/glm-asr` | GLM ASR |
| 任务查询 | `/v3/async/task-result?task_id=xxx` | - |
---
## 执行模板
### 文生图
```bash
curl -X POST "https://api.ppio.com/v3/seedream-5.0-lite" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "描述"}'
```
### 图片编辑
```bash
curl -X POST "https://api.ppio.com/v3/seedream-5.0-lite" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "编辑指令", "reference_images": ["图片URL"]}'
```
### 文生视频
```bash
curl -X POST "https://api.ppio.com/v3/async/vidu-q3-pro-t2v" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "描述", "duration": 4}'
```
### 图生视频
```bash
curl -X POST "https://api.ppio.com/v3/async/vidu-q3-pro-i2v" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "动作描述", "images": ["图片URL"]}'
```
### TTS
```bash
curl -X POST "https://api.ppio.com/v3/async/minimax-speech-2.8-turbo" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"text": "要转换的文字",
"voice_setting": {"voice_id": "male-qn-qingse", "speed": 1.0},
"audio_setting": {"format": "mp3"}
}'
```
**可用声音:**
- 男声:`male-qn-qingse`(青涩)、`male-qn-jingying`(精英)
- 女声:`female-shaonv`(少女)、`female-yujie`(御姐)
### STT
```bash
curl -X POST "https://api.ppio.com/v3/glm-asr" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"file": "音频URL或Base64"}'
```
### 任务结果查询
```bash
curl "https://api.ppio.com/v3/async/task-result?task_id=$TASK_ID" \
-H "Authorization: Bearer $API_KEY"
```
**状态:** `TASK_STATUS_QUEUED` → `TASK_STATUS_PROCESSING` → `TASK_STATUS_SUCCEED`
---
## 错误处理
| 错误码 | 含义 | 处理 |
|--------|------|------|
| 401 | Key 无效 | 检查配置 |
| 402 | 余额不足 | https://ppio.com/billing 充值 |
| 429 | 请求过快 | 等待重试 |
## 定价
https://ppio.com/pricing
ContextWeave one-shot architecture diagram generation
A script-based skill that generates or edits images with Nano Banana Pro.
Runs the whisper command line to transcribe audio locally.
A skill that generates and edits diagrams through an MCP server.
Design intelligence and implementation guidance for polished interfaces, from layout and flows to tokens and frontend changes.
An AI video generation and orchestration skill across several model APIs.
Pull a frame out of a video with ffmpeg, or produce a thumbnail for a quick look.
Turn audio files into text with local Whisper, running fully offline once the model is downloaded.
Data month: 2026-09 · Downloads, stars and installs are aggregated monthly from public skill registries (ClawHub, SkillHub). See methodology.