agentleFS
lorena-bordonaba-paulorena-bordonaba-pau/youtube-agent/.claude/skills/script-writing/SKILL.md

script-writing

Skill2 starsChanged 47 days agoMITView on GitHub

Write video scripts in the channel's real voice with a validated retention structure. Triggered by "write the script", "give me the hook", "structure this video", "how do I open this one". In Spanish, "escribe el guion", "dame el hook", "estructura este vídeo", "cómo abro este vídeo".

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If it is populated (more than 150 words, with verbatim phrases), use it and skip to step 3.

    line 18
  2. Distinctive:

    If it is empty or generic, build it before writing anything:

    line 20
  3. Very specific:

    Read memory/sop/scripting_sop.md.

    line 32
  4. Distinctive:

    Look at where the audience dropped in that video and do not repeat the pattern.

    line 46
  5. Distinctive:

    Write, applying the SOP and the voice from step 1.

    line 50

“Common” is measured: kept by at least 6 in 10 of the 3 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

Script

WHEN

Writing or restructuring a script, designing the first-30-seconds hook, placing re-hooks, reviewing a structure already written.

EXECUTION ORDER

  1. Check the voice profile Read memory/voice_profile.md.

    • If it is populated (more than 150 words, with verbatim phrases), use it and skip to step 3.
    • If it is empty or generic, build it before writing anything:
      python3 tools/yt_top_videos.py --days 180 --by-retention --limit 5
      python3 tools/yt_transcript.py --video ID --plain    # the best 3
      Analyse the real fillers, openings, transitions and closings, and write memory/voice_profile.md with verbatim quotes.

    Inventing the voice when transcripts are available is the error this step prevents.

  2. The SOP has structural primacy Read memory/sop/scripting_sop.md. It is the source of truth for structure. If a formula exists there, the outliers and keywords from the next steps inform the TOPIC, they never rewrite the structure.
  3. Validate the topic
    python3 tools/yt_outliers_channels.py --min-ratio 2.0
    python3 tools/kw_research.py --kw "main topic; variant"
    The outliers say which hooks are earning views right now. The keyword score is heuristic: it orders options, it does not assert demand.
  4. Verify the hook against real retention — if a comparable video exists python3 tools/yt_retention.py --video ID Look at where the audience dropped in that video and do not repeat the pattern. This turns the SOP into something verifiable rather than an inherited convention.
  5. Write, applying the SOP and the voice from step 1.

SOURCES

Step 1 youtube_api (transcripts). Step 3 derived + heuristic. Step 4 youtube_api. When citing the keyword score, declare it.

OUTPUT

Whatever was asked for: a hook broken down by seconds, a summary outline, or a full script. Summary by default; the full script only on confirmation.

Each block declares its job (promise, proof, plan, re-hook) so it can be measured afterwards with yt_retention.py.

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.