agentleFS
Sign inStart free
scrollmarkscrollmark/social-skills/skills/editor-captions/SKILL.md

editor-captions

Skill19 starsChanged 29 days agoMITView on GitHub

Use when adding or restyling captions in the Scrollmark editor, one word at a time (the default for "Add captions" and "Cut this for TikTok" on a talking-head video) or in short three-word chunks.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Never phrase (8-word lines), never phrase-plus-word (an extra word track), and never captions_generate: it makes sequential chunks without …

    line 17
  2. Distinctive:

    After the recut, or on a timeline that already has the keep list:

    line 21
  3. Very specific:

    Do not use tiktok, creator-classic, sticker-box, cinematic, yellow-punch, meme-loud, or active-word unless the user names that look.

    line 31
  4. Very specific:

    If captions already exist with role word, captions_resync.

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Editor Captions

Requires

A running Scrollmark editor whose commands you can call: the SocialGPT editor chat, or scrollmark mcp from @scrollmark/cli for a coding agent. Without it the steps below can't run; the kit-by-vibe list is still a fair guide to choosing a caption look in any editor.

Pick the rhythm

  • One word at a time (default): captions.applyRhythm with mode: "word".
  • Three-word chunks, when the user wants chunks rather than single words: captions.applyRhythm with mode: "sequential" and wordsPerCaption: 3.

Never phrase (8-word lines), never phrase-plus-word (an extra word track), and never captions_generate: it makes sequential chunks without the rhythm you chose.

Steps

After the recut, or on a timeline that already has the keep list:

  1. The rhythm op above, then captions.applyPreset with kitId, size and placement: as two ops in one timeline_applyPlan, or as captions_applyRhythm and captions_applyPreset outside a plan.

  2. Pick one kitId from this set. Choose from the footage (transcript, setting, what they are selling). Do not call captions_listPresets just to browse.

    • handwritten — casual talking-head, lifestyle, personal story, home or messy setting
    • soft-promo — product, beauty, UGC, "check this out" energy
    • podcast-plate — interview, seated conversation, darker or serious talk
    • clean-minimal — default if unsure; explainer, professional, clean talking-head

    Do not use tiktok, creator-classic, sticker-box, cinematic, yellow-punch, meme-loud, or active-word unless the user names that look.

  3. Size: s for word captions, since at m long words clip on a vertical video; m or l only when the user asks for bigger captions.
  4. Placing for a platform (the user named one, or the editor's safe zone shows one): pass the same platform to captions.applyRhythm and to captions.applyPreset. The preset recomputes where captions sit, so leaving it out of either call puts them back under the platform's caption block. Then check that platform with qc_run (checks: ["document"]) and fix what it finds with qc_applyFix.
  5. If captions already exist with role word, captions_resync. It keeps their kit, colour and size, so do not apply a preset afterwards unless the user asks for a different look.

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.

Check your setup, free