How to batch format douyin content for creator teams
Creator teams batch-format douyin posts: Token Forge sets the palette, Brand Canvas and Film Lens make covers, Pixel Forge resizes, Text to Speech adds audio.
Skills in this bundle
Text to Speech
Convert text into high-quality MP3 audio with selectable voices and speed, running locally with no cloud API (powered by the open-source Kokoro model). Use it to turn articles or manuals into audio for videos or podcasts.
Token Forge
Extract colors, typography, spacing and other design tokens from design files and export to developer-friendly formats like CSS variables, JSON, or SCSS.
Image Shrink
Compress JPEG, PNG, WebP, and GIF images locally with adjustable quality to minimize file size without any cloud upload.
Pixel Forge
Deterministic batch image processing: resize, format-convert, compress, crop, thumbnail, rotate, flip, grayscale, and metadata extraction (Pillow).
Brand Canvas
Generate consistent brand marketing images from text prompts, supporting social media posts, posters, and promotional materials with customizable styles.
Creator teams format douyin content in bulk by chaining six skills inside Nexus Runner on their own machine: Token Forge fixes the palette and type scale, Brand Canvas and Film Lens supply covers and photography, Pixel Forge standardizes vertical sizes and formats, Image Shrink compresses the files, and Text to Speech turns scripts into MP3. One pass leaves a full episode of upload-ready images and audio in a local folder, with nothing sent off the machine.
- One pass covers five stages — set the spec, source assets, standardize sizes, compress, voice the script — for 36 Nex across six skills.
- Token Forge extracts colors, typography and spacing from design files and exports CSS variables, JSON or SCSS as the account's shared visual spec.
- Pixel Forge batch-resizes, crops, converts, thumbnails, rotates and flips; Image Shrink compresses JPEG, PNG, WebP and GIF locally at adjustable quality.
- Text to Speech runs the open-source Kokoro model on your PC to produce MP3 with selectable voices and speed, with no cloud API call.
- All six run on Nexus Runner 1.0.10 (Windows 10/11 x64 only); output lands in a local folder and the data never leaves the machine.
What batch-formatting douyin content means
Batch-formatting douyin content means producing every cover, still, thumbnail and voiceover an episode needs in one pass against a single visual spec, instead of handling each item by hand. On MCP Nexus six locally callable skills cover that work: Token Forge holds the visual spec, Brand Canvas and Film Lens source the imagery, Pixel Forge and Image Shrink handle dimensions and file size, and Text to Speech handles narration. All six are single-call skills (the other type is agent, for multi-step orchestration), each priced per run in Nex, writing straight to a folder on your machine for your team to upload.
How the six skills chain into one pipeline
Three ways to do the same job
| MCP Nexus | Manual work | Generic cloud AI | |
|---|---|---|---|
| Where assets live | Skills run in Nexus Runner 1.0.10 on your PC; images and MP3 land in a local folder and the data never leaves it | Files stay local, but every step is opened, exported and renamed by hand | Assets must be uploaded to a cloud server before anything happens |
| Batch handling | One call processes a whole set, and the six stages run back to back | Resize one, compress one, record one, repeat | Mostly one item per conversation turn; batching is something you assemble yourself |
| Visual consistency | The palette and type table from Token Forge carries through cover production and is reusable next episode | Relies on people remembering the spec; drifts across staff and episodes | Style is re-described each time, so output is hard to reproduce |
| How you pay | Per run in Nex — 36 Nex for one full pass of six skills; the free tier gives 200 Nex a month, reset at the start of each month | Priced in staff hours | Priced by subscription seats and usage allowance |
| What you get back | Upload-ready image and MP3 files on disk, with naming and folders under your control | A finished batch that still has to be sorted and renamed | Items downloaded one by one, then filed by hand |
Real use cases
FAQ
How do I resize douyin covers to one size in bulk?
Use Pixel Forge for batch resizing and cropping. Built on Pillow, it handles resize, format conversion, crop, thumbnail, rotate, flip, grayscale and metadata extraction deterministically, so a whole set becomes one vertical spec in a single run before Image Shrink compresses it.
Can narration scripts be converted to voiceover in bulk?
Yes, with Text to Speech. It runs the open-source Kokoro model locally to produce high-quality MP3 with selectable voices and speed, no cloud API involved, which suits video voiceover or audio versions of articles and manuals.
Do these skills upload my assets to the cloud?
No. They run in Nexus Runner 1.0.10 on your own PC — Image Shrink, Pixel Forge and Text to Speech all process locally, and output is written to a folder on your machine. Runner supports Windows 10/11 x64 only.
How much Nex does a full pass cost?
36 Nex for all six: Text to Speech 10, Brand Canvas 10, Token Forge 5, Film Lens 5, Pixel Forge 3, Image Shrink 3. The free tier's 200 Nex a month (reset at the start of each month) covers five full passes; Pro is ¥49/month with 3000 Nex and Enterprise ¥299/month with 50000 Nex.
Can it publish to my douyin account for me?
No. The pipeline produces local files — covers, stills, thumbnails, MP3 audio and the token table — and publishing stays with your team. These skills do not operate third-party platform accounts and do not drive the screen.