agentleFS
alisadiq-aialisadiq-ai/favstash-ai-video-editing-skills/skills/1-research/find-ideas/SKILL.md

find-ideas

Skill75 starsChanged 40 days agoApache-2.0View on GitHub

Find what to make next: proven reels to adapt, outliers in a niche, saved inspiration and the creator's own results. Use before scripting, when asked what to post, to judge whether a reference is proven, or to turn saved reels into a short list of ideas.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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. Distinctive:

    Start every reel from evidence: a reference that already worked for someone with a similar audience, or a result from the creator's own …

    line 8
  2. Distinctive:

    Call a reference proven only against its own creator's baseline: views at least 3× the median of a comparable recent sample of their posts.

    line 28
  3. Very specific:

    Look at the reference itself when the treatment matters: the first frame, the cuts, the captions and where sounds land.

    line 48

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Find ideas

Start every reel from evidence: a reference that already worked for someone with a similar audience, or a result from the creator's own posts. The output is a short list of ideas, each with its reference, why it's worth copying and the angle that's true for this creator. Hand the chosen one to write-script.

Where ideas come from

  • References the creator supplies: a reel URL, a saved post or a screen recording.
  • The creator's saved inspiration: with FavStash connected, search the stash by topic, hook pattern or audience problem, and read saved transcripts and on-screen text instead of guessing them. Saving a new URL keeps it with the rest.
  • Niches and competitors: recent posts from creators the audience already follows. Compare each post with that creator's own baseline, not with other accounts.
  • The creator's results: with FavStash analytics, their own outliers and their weakest openings are the best guide to the next test.

Without FavStash, work from supplied URLs and transcripts; everything below still applies.

Is the reference proven?

Call a reference proven only against its own creator's baseline: views at least 3× the median of a comparable recent sample of their posts. Record the views, publish date, capture date and sample size. Without a baseline it is a format reference: useful for structure, not proof that the hook works.

  • A high comment count on a comment-to-DM post shows the gate converted, not that the hook earned reach.
  • Performance shows association. Topic, timing and distribution stay confounded.
  • Saved content is taste and structure, never factual proof for the creator's claims.

What to record for each idea

Idea: one line, in the viewer's words
Reference: URL, creator, views vs their median (proven | format | original)
Why it worked: the hook lever, the first frame, the payoff
True for this creator: what they can honestly say or show; what must change
Proof available: footage, screen capture, real numbers (or what's missing)

Look at the reference itself when the treatment matters: the first frame, the cuts, the captions and where sounds land. A transcript alone doesn't show the edit; edit-video's reference helper samples frames from a URL into an analysis copy.

Give the creator three to five ideas, strongest first, and say which one you'd make. Rights don't transfer with an idea: adapt the hook, structure and pacing, and make the video from the creator's own footage.

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.