agentleFS
uplikauplika/uplika-plugin/skills/yt-shorts/SKILL.md

yt-shorts

Skill3 starsChanged 35 days agoMITView on GitHub

Find the Shorts inside a long YouTube video from its transcript, write a new first line for each, and upload them. For 'cut this into shorts', clips.

7 other projects use a version of this, 5 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read ~/.uplika/<workspace-slug>/voice.md if it exists, for the new first lines.

    line 20
  2. Very specific:

    Read the transcript and find spans of roughly 20-55 seconds (the original pack's rule of thumb for a Short that holds.

    line 24
  3. Very specific:

    Read YouTube's shorts rule in list_platforms: a vertical video inside that length goes up as a Short, and a custom thumbnail does not apply …

    line 45
  4. Very specific:

    If a call answers workspace_required, show the workspaces it lists, ask which one, and pass workspaceId.

    line 48

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Adapted from Jakeschincariol/youtube-agent-skill (MIT). Not affiliated with the original author.

yt-shorts

A Short cut out of a long video is not a clip of the best moment. It is a moment that survives without the video around it, which is a much smaller set. Use it for "cut this into shorts", "clip this", "repurpose this video", "what should I clip".

The input is the long video's transcript (srt, vtt or whisper json). When a shell is available and the video is on YouTube, this saves its automatic captions (use the video's language code):

yt-dlp --skip-download --write-auto-subs --sub-langs ko --convert-subs srt "<video link>"

Before you write

Read ~/.uplika/<workspace-slug>/voice.md if it exists, for the new first lines. If it does not exist, create it with the voice-profile skill, or ask its three questions (tone, ending style, words to avoid) and use the answers for this session.

Picking

Read the transcript and find spans of roughly 20-55 seconds (the original pack's rule of thumb for a Short that holds; YouTube's own Shorts rule is in list_platforms) where all three are true:

  1. It opens on a complete thought. If the first sentence needs the previous minute, it is not a Short.
  2. There is a turn in it: a claim, then something that complicates or proves it.
  3. It ends on a line, not a trail-off.

Rank the candidates and show the person the top five with their timecodes and first line, so they can reject one without reading the whole transcript.

Writing each one

  • A NEW first line. The long video's line assumes context this viewer does not have. Write the replacement and run it through ../yt-script/scripts/hookscore.py (a Korean line gets a rank-only score marked Korean: heuristic, not calibrated).
  • On-screen text for the first two seconds, different words from the spoken line.
  • A loop point: what the last line sets up so the first line answers it.
  • Vertical framing note: what gets cropped out of a 16:9 frame and whether that matters.

With Uplika connected

Uplika cannot cut video: the person cuts each Short in their editor as a vertical clip. Then, for each Short:

  1. select_channels with scope youtube gives the account id to post to, and skipped with a reconnect link if the connection needs it. Nothing connected: https://uplika.com/dashboard/connections.
  2. The file: if it has a public https address, media_from_url; if it is on the person's computer or phone, upload it the way the yt-upload skill does.
  3. Read YouTube's shorts rule in list_platforms: a vertical video inside that length goes up as a Short, and a custom thumbnail does not apply to it.
  4. Show the person the title (options.youtube.title; the new first line is a good start), the description, the privacy (options.youtube.privacyStatus; left out, it goes up private), whether it is made for kids (options.youtube.madeForKids: ask, and send it only from their answer), and when it goes out. Only after they approve, call publish with the account id, the media id, wait: true, and scheduledAt with a time zone offset if it should go out later.
  5. get_post gives the status, the link and any warning (for example the privacy YouTube actually applied). Read the warning out.
  6. If a call answers workspace_required, show the workspaces it lists, ask which one, and pass workspaceId. Errors: get_help with the code.

Without Uplika

The person cuts and uploads the Shorts themselves: hand them the titles, first lines, on-screen text and timecodes.

The gate

Nothing goes out until the person says yes. Every output ends in a block they can copy, and the last line of every run is the question: ship it, or change it? With Uplika connected, "ship it" means you upload the approved Shorts with publish.

Tools

  • select_channels: the YouTube account to post to
  • list_platforms: YouTube's Shorts rule
  • media_from_url: attach a Short that already has a public https address
  • publish: upload an approved Short, now or at scheduledAt
  • get_post: status, link and warnings after the upload
  • get_help: the answer for an error code

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.