agentleFS
Sign inStart free
cgoinglovecgoinglove/thursday-agent/skills/media-digest/SKILL.md

media-digest

Skill56 starsChanged 47 days agoMITView on GitHub

Reads anything long — a video, podcast, talk, article, PDF or a recording of their own — and hands it back short. Use it when the user points at one and asks what is in it, wants a meeting or lecture they recorded written up, or asks which videos on a topic are worth watching.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. cgoinglove/thursday★ 56100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository12 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    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. Very specific:

    Every script is node <this skill's dir>/scripts/<name>.mjs, needs no key and prints a few lines.

    line 8

≈ “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

Media digest

Every script is node <this skill's dir>/scripts/<name>.mjs, needs no key and prints a few lines; the text itself goes to a file. Chain them with && in one bash call.

S=<this skill's dir>/scripts
node $S/yt.mjs transcript <url|id>... --out <scratch> [--lang de]   # <id>.txt + <id>.json per video
node $S/yt.mjs search "<query>" [--within hour|day|week|month|year] [--length short|medium|long] [--sort relevance|views|date] [--max 20] --out <scratch>/rows.json
node $S/text.mjs <article url | pdf url | file.pdf> --out <scratch>/<name>.txt
node $S/part.mjs <file> <n>                                         # part n of a long file, one read
node $S/audio.mjs cut <url|file> --out <scratch>/audio [--minutes 10]   # no captions: pieces to transcribe
node $S/digest.mjs <digest.json> --name <page>                      # the page, in your artifacts folder
node $THURSDAY_SKILLS/artifact/scripts/document.mjs pdf <page path>   # a PDF beside it, when asked for one

Two facts decide the work:

  • One bash call shows only so much of what it prints; the rest is cut. So every file these scripts write is split into ## Part N blocks that each fit one read, and the head they print says how many. One part per bash call — two in one call are cut — and the calls for several parts sent together in one step. A question about one thing is grep -n -i on the file first, then only the parts that hit; a summary reads every part. How: references/reading.md.
  • A timestamp is only worth giving when it comes from the file. Each transcript line starts with [m:ss]; cite that time, never one worked out. A ~ marks a time estimated from where the words fall in a piece of audio; cite it with the ~, which still opens the source at that second.

YouTube is read with yt-dlp, which the first yt.mjs or audio.mjs call fetches into projects/ once (~50 MB); after that a video takes a few seconds.

references/page.md is the page's JSON and how an answer is shaped; references/no-captions.md is a video, podcast or recording with no transcript. A meeting or lecture they recorded is written up as notes — what was decided, who does what and by when, each point with its [m:ss] — in the artifact skill's templates/document/notes.md, not as a digest page.

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