According to AlphaMoat, as of 2026-09, Pipixia Drama Producer has 1,466 downloads and 0 stars — ranked #4,785 of 5,078 AI skills overall, and #475 of 512 in Design & Media.
| Platform | Installs | Downloads | Stars | Comments |
|---|---|---|---|---|
| ClawHub ↗ | 31 | 1.2K | - | - |
| SkillHub ↗ | 284 | 1.5K | - | — |
Platforms count differently and update at different times.
A six-step pipeline: I2V image-to-video (must use reference_type="first_frame") → normalize and trim the first second (the first second of each image-to-video clip is a still frame of the source image and must be trimmed) → concat editing (insert 1.5s reaction shots after key lines) → node-edge-tts dubbing (TTS dialogue must not exceed its corresponding video clip; if it does, speed it up with atempo=1.1~1.2 speed it up) → mix_audio.py timeline mixing with BGM (Kevin MacLeod CC BY 3.0 recommended) → extract a cover and send via send_video.sh/send_audio.sh to a Feishu chat. Requires FEISHU_APP_ID/FEISHU_APP_SECRET and the im:message:send, im:file, im:image permissions; keep versioned outputs (v2/v3). Card safety: the release security verdict is suspicious; send_video.sh interpolates media paths into python3 -c — process only trusted paths or patch it first; replace shared /tmp with mktemp private dirs in shared environments; require explicit operator confirmation before sending.
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 | 1.5K | — | - |
| 2026-08 | 1.2K | +7.5% | - |
| 2026-07 | 1.1K | +16.6% | - |
| 2026-06 | 984 | +9.0% | - |
| 2026-05 | 903 | — | - |
| 2026-03 | 549 | — | - |
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: pipixia-drama-producer description: 皮皮虾职场短剧全流程制作技能。用于为「皮皮虾」(机械龙虾AI-bot)职场短剧生成镜头视频、剪辑成片、配音配乐并发布到飞书群。完整流程:图生视频(I2V) → ffmpeg规范化+剪辑 → TTS配音 → BGM混音 → 飞书媒体消息发送。当用户提到制作皮皮虾短剧、生成新镜头、剪辑视频、配音配乐、或将视频/音频发送到飞书群时激活。 --- # 皮皮虾短剧制作技能 完整的职场动物短剧制作流程,从剧本到飞书发布。 ## 环境变量配置
使用前请设置以下环境变量:
| 变量名 | 必填 | 说明 |
|--------|------|------|
| `FEISHU_APP_ID` | ✅ | 飞书应用 App ID(发送消息用) |
| `FEISHU_APP_SECRET` | ✅ | 飞书应用 App Secret |
| `FFMPEG` | 可选 | ffmpeg 路径(默认从 PATH 查找) |
| `FFPROBE` | 可选 | ffprobe 路径(默认从 PATH 查找) |
| `EDGE_TTS` | 可选 | edge-tts 路径(默认从 PATH 查找 `node-edge-tts`) |
**飞书应用权限要求**:`im:message:send`、`im:file`、`im:image`
## 角色与剧本参考
详见 `references/drama-reference.md`,包含:
- 角色设定与prompt模板
- ffmpeg关键命令速查
- TTS音色对照表
- 飞书API速查
## 核心工作流
### 第一步:生成镜头(I2V图生视频)
```python
# 必须用 reference_type="first_frame",不要用 "subject"
gen_videos([{
"image_file": "/workspace/lobster_robot.png",
"output_file": "/workspace/shot.mp4",
"prompt": "cute orange lobster robot [动作], ...",
"reference_type": "first_frame"
}])
```
### 第二步:规范化 + 砍首秒
每个图生视频第一秒是原图静帧,必须裁掉:
```bash
python3 scripts/normalize_and_trim.py input.mp4 output_trim.mp4 --trim 1
```
### 第三步:剪辑合并
```bash
# 写concat列表,用Python避免shell语法问题
python3 -c "
clips = ['clip1_trim.mp4', 'clip2_trim.mp4', ...]
with open('/tmp/concat.txt', 'w') as f:
for c in clips: f.write(f\"file '{c}'\n\")
"
ffmpeg -y -f concat -safe 0 -i /tmp/concat.txt -c copy output.mp4
```
**剪辑技巧**:
- 在关键台词后插入1.5s反应镜头(增加喜剧节奏)
- 砍掉所有图生视频开头静帧(-ss 1)
### 第四步:TTS配音
```bash
node-edge-tts -t "台词内容" -f output.mp3 -v zh-CN-YunxiaNeural -l zh-CN
```
⚠️ TTS台词时长不能超过对应视频片段时长,过长需用 `atempo=1.1~1.2` 加速:
```bash
ffmpeg -y -i input.mp3 -filter:a "atempo=1.15" output.mp3
```
### 第五步:混音(配音+配乐)
使用 `mix_audio.py` 将多条TTS台词按时间轴叠加,并加入BGM:
```bash
python3 scripts/mix_audio.py \
--video drama_cut.mp4 \
--bgm bgm_sneaky.mp3 \
--bgm-volume 0.25 \
--output drama_dubbed.mp4 \
--lines "0.3:tts_line1.mp3" "5.1:tts_line2.mp3" "12.0:tts_line3.mp3"
```
BGM推荐下载地址(Kevin MacLeod CC BY 3.0):
```bash
curl -sL "https://incompetech.com/music/royalty-free/mp3-royaltyfree/Sneaky%20Snitch.mp3" -o bgm.mp3
```
### 第六步:发送到飞书
```bash
# 提取封面
ffmpeg -y -i video.mp4 -ss 00:00:00.5 -frames:v 1 -update 1 cover.jpg
# 发送视频(媒体气泡),chat_id 为目标群组 ID
bash scripts/send_video.sh video.mp4 cover.jpg <chat_id>
# 发送音频语音消息
bash scripts/send_audio.sh "要播报的文字" <chat_id> "zh-CN-YunxiaNeural"
```
## 依赖工具
| 工具 | 环境变量 | 默认值 |
|------|----------|--------|
| ffmpeg | `FFMPEG` | `ffmpeg`(PATH) |
| ffprobe | `FFPROBE` | `ffprobe`(PATH) |
| edge-tts | `EDGE_TTS` | `node-edge-tts`(PATH) |
| 飞书凭证 | `FEISHU_APP_ID` / `FEISHU_APP_SECRET` | 无(必填) |
## 版本管理惯例
每次重要修改用v2/v3...命名输出文件,保留历史版本方便回退:
- `drama_ep1_v1.mp4` → `drama_ep1_v2.mp4` → ...
- 在工作目录记录每个版本的剪辑结构
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.