S
StratosHouse/altocue
The AI edit suite for coding agents
View on GitHub ↗Link checked 2026-11-14 · HTTP 200 · opens the vendor's own site
Official descriptionThe AI edit suite for coding agents. Raw footage in, finished video out.
Install & config
Client config (built from the install command this repo gives)
{
"mcpServers": {
"stratoshouse-altocue": {
"command": "uvx",
"args": ["altocue", "mcp"]
}
}
}Parsed from the README we fetched on 2026-11-13; we have not executed it.
command/args come from the install command the repository states. When it is ambiguous, the project README wins.
VS Code (.vscode/mcp.json - the key is servers)
{
"servers": {
"stratoshouse-altocue": {
"command": "uvx",
"args": ["altocue", "mcp"]
}
}
}Same server, but VS Code keeps it under <code>servers</code>.
Claude Code (command line)
claude mcp add --transport stdio stratoshouse-altocue -- uvx altocue mcp
Claude Code does not read a config file; add the server with this command.
Key names differ by client: Clients & CLI
- Star count
- 0 ★ · StratosHouse/altocue · read from the GitHub search API · (checked 2026-11-13 20:00:50)
- Source
- read directly from its GitHub page · type: open-source project
- Harvest history
- first seen 2026-11-14 · last seen 2026-11-14 · seen 1× · listed by 1 sources
- Score
- score 103 · 1 sources · repo
- Install
uvx altocue mcp
FAQ (answers come from our records)
How do I install it?
The repository provides this install command:
uvx altocue mcp. It comes from the repository itself; we cannot guarantee it runs as-is in your client.Who is it for?
We filed it under these tags: Agent 技能. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository
StratosHouse/altocue；first catalogued 2026-11-14；last seen 2026-11-14。There is no official certification or quality rating here — we record sources and dates only.Altocue
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
The AI edit suite for coding agents.
A full open video editing suite your agent drives. It cuts, cleans, grades, captions, animates, adds sound and delivers, and it makes films with no footage at all. Then a separate Editor in Chief has to sign off.
The presenter is AI-generated with Higgsfield. The mistakes are planted. The edit is real.
> Making a launch video? There is a dedicated feature for it, and a skill that runs it from the brief to the finished film. It needs no footage, no filming and no keys:
> bash
> altocue launch my-launch --product "Name"
> Or ask your agent for a launch video. The altocue-launch skill picks it up. See Films with no footage.
## What it does
You hand your coding agent a messy recording, a long talk or just a brief. Altocue gives it the eyes, ears and tools to finish the job.
Reframing that follows. A tall crop that trails someone walking and cuts with the camera, and names the seconds it saw no face. With speaker labels it can also cut to whoever is talking in a wide shot of two, which has not yet worked on real footage. A still crop stays wherever it already works.
Motion tracking. Blur, pixelate, spotlight, a label or a sticker that stays on a face or on a box you draw, followed through the film on your machine.
Cutting. Word-timed transcripts with fillers kept, so cuts land on real speech. Each removal has a reason you can overrule. Or edit the script and get a frame-safe cut.
Sound. Noise and hum out, music under the voice taken off when it is measured and Demucs is installed, each phrase levelled, room tone in the holes, mastered to a measured loudness and true peak.
Colour. HDR to SDR, white balance and exposure read from the face, film looks or your own LUT, with skin protected.
Captions. Word by word, never across a cut, only words really said. Speakers can be named or coloured.
Motion graphics. 151 original formats in 12 groups, timed to speech and placed clear of faces. The Studio collection adds smooth typography, depth, linked handoffs, editing diagrams and endings.
Copyable motion recipes. Every built-in graphic includes an agent prompt, a native example and optional sound notes. Preview the full library and adapt it to your words.
Music and effects. Sound you are allowed to use, with the licence and credit shown before anything downloads. Single CC0 sounds are indexed from open packs by category, from clicks and whooshes to glitches and risers, and paid stores are plain links.
Product UI inserts. Faux product screens built with Stageshow, filmed one fixed frame at a time into an insert that lands on the spoken word, and labelled as a mockup.
Films with no footage. Launch films, explainers, demos, ad cards and more, drawn in code from a storyboard.
Long video to shorts. It finds complete thoughts in your own recordings and cuts each into its own project.
Deliverables. Subtitles, chapters, thumbnails, stems, a version with no music for an app's own song library, and a version for each platform, with the master untouched.
Data checks. It measures the finished film and learns from your own results. It predicts nothing.
The Editor in Chief. It does not mark its own homework. Fact gates in code, then a separate reviewer, then code decides SHIP or FIX.
A full open video editing suite your agent drives. It cuts, cleans, grades, captions, animates, adds sound and delivers, and it makes films with no footage at all. Then a separate Editor in Chief has to sign off.
The presenter is AI-generated with Higgsfield. The mistakes are planted. The edit is real.
> Making a launch video? There is a dedicated feature for it, and a skill that runs it from the brief to the finished film. It needs no footage, no filming and no keys:
> bash
> altocue launch my-launch --product "Name"
> Or ask your agent for a launch video. The altocue-launch skill picks it up. See Films with no footage.
## What it does
You hand your coding agent a messy recording, a long talk or just a brief. Altocue gives it the eyes, ears and tools to finish the job.
Reframing that follows. A tall crop that trails someone walking and cuts with the camera, and names the seconds it saw no face. With speaker labels it can also cut to whoever is talking in a wide shot of two, which has not yet worked on real footage. A still crop stays wherever it already works.
Motion tracking. Blur, pixelate, spotlight, a label or a sticker that stays on a face or on a box you draw, followed through the film on your machine.
Cutting. Word-timed transcripts with fillers kept, so cuts land on real speech. Each removal has a reason you can overrule. Or edit the script and get a frame-safe cut.
Sound. Noise and hum out, music under the voice taken off when it is measured and Demucs is installed, each phrase levelled, room tone in the holes, mastered to a measured loudness and true peak.
Colour. HDR to SDR, white balance and exposure read from the face, film looks or your own LUT, with skin protected.
Captions. Word by word, never across a cut, only words really said. Speakers can be named or coloured.
Motion graphics. 151 original formats in 12 groups, timed to speech and placed clear of faces. The Studio collection adds smooth typography, depth, linked handoffs, editing diagrams and endings.
Copyable motion recipes. Every built-in graphic includes an agent prompt, a native example and optional sound notes. Preview the full library and adapt it to your words.
Music and effects. Sound you are allowed to use, with the licence and credit shown before anything downloads. Single CC0 sounds are indexed from open packs by category, from clicks and whooshes to glitches and risers, and paid stores are plain links.
Product UI inserts. Faux product screens built with Stageshow, filmed one fixed frame at a time into an insert that lands on the spoken word, and labelled as a mockup.
Films with no footage. Launch films, explainers, demos, ad cards and more, drawn in code from a storyboard.
Long video to shorts. It finds complete thoughts in your own recordings and cuts each into its own project.
Deliverables. Subtitles, chapters, thumbnails, stems, a version with no music for an app's own song library, and a version for each platform, with the master untouched.
Data checks. It measures the finished film and learns from your own results. It predicts nothing.
The Editor in Chief. It does not mark its own homework. Fact gates in code, then a separate reviewer, then code decides SHIP or FIX.
Listed badge (optional)
Copy this line into the README if you want the listing mark. Optional, and it does not affect being listed.
[](https://anclum.com/en/s/stratoshouse-altocue/)
- Source
- github · https://github.com/StratosHouse/altocue
- Browse by field
- Agent skills
- added
- today
More like this
Filed under
Every tag opens the full list for that filter. Counts are the live catalogue figures.