Publish-Assistant

Connect OpenRouter

Publish-Assistant uses your own OpenRouter key for transcription, titles, summaries, tags and thumbnails. The key stays with you.

or paste a key you already have:

How to get an OpenRouter key

Drop a video here

or choose files. Several files run one at a time.

Generate

Titles Summary Tags Thumbs Thumbnail shape:

Queue

Results

Chosen title

Summary

Thumbnail

Files

Select a title (best first)

Or edit / write your own

Review thumbnail concepts

Edit any concept, untick the ones you do not want, then render.

Select a thumbnail

Settings

OpenRouter key

Summary / titles / tags model Titles per video (5 to 15)

Thumbnail model Thumbnails per video (1 to 6)

Speech model Identify speakers (Deepgram, about $0.25 per hour of audio) Custom spelling rules (one per line: Correct = wrong one, wrong two)

Each prompt has two parts: your own instructions (tone, audience, brand, anything extra) and the rules the app always adds after them. The rules carry the format the app relies on, so they cannot be edited. The transcript is added last.

Title prompt: your instructions

Rules the app always adds

Summary prompt: your instructions

Rules the app always adds

Backlinks (inserted before *Key Points*; left out until you replace Your-URL-Here) Tags prompt: your instructions

Rules the app always adds

Brand colors

Used by both thumbnail prompts. Pick with the swatch or paste a hex value like #F2B33D. Leave empty to let the model choose.

Primary

Secondary 1

Secondary 2

Background

Thumbnail prompt (16:9) Shorts thumbnail prompt (9:16)

Your face for the thumbnails. One photo is picked at random per video (never the same one twice in a row).

Upload defaults. Uploading itself arrives once Google approves the web sign-in.

Visibility

Connect OpenRouter

Generate

Queue

Results

Chosen title Copy Choose another

Summary Copy

Tags Copy

Thumbnail Get thumbs Download

Files

Chosen title

Summary

Tags

Thumbnail