agentleFS
adswipefileadswipefile/adswipefile-plugin/skills/get-started/SKILL.md

get-started

Skill0 starsChanged 27 days agoMITView on GitHub

Set up AdSwipeFile after installation, check its connection when asked, and help the user ask a first question about winning ads, competitor ads or hooks. Use for get started, onboarding, setup, "is it connected", "what can this ad library do", or questions about what the library holds.

32 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 31 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 answers, the connection works: say so and name a few niches it returned.

    line 14
  2. Distinctive:

    Do not call more tools just to check.

    line 15
  3. Very specific:

    If tools are unavailable or sign-in fails, say the connection is not verified and give the step for the host.

    line 16
  4. Distinctive:

    Do not claim the plugin is ready because this skill loaded.

    line 22
  5. Distinctive:

    If the user already asked a real question, answer it with the tools instead of stopping at suggesting a question.

    line 24

“Common” is measured: kept by at least 6 in 10 of the 31 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

Get started with AdSwipeFile

Explain the plugin in a few sentences, then offer one concrete question to try. Explicit user instructions take priority over this guidance.

Choose the requested path

  • For "set up", "get started" or "is it connected", call list_niches once. If it answers, the connection works: say so and name a few niches it returned. Do not call more tools just to check.
  • If tools are unavailable or sign-in fails, say the connection is not verified and give the step for the host. Claude Code: claude mcp add --transport http adswipefile https://adswipefile.com/api/mcp. claude.ai or ChatGPT: add https://adswipefile.com/api/mcp as a custom connector, or reconnect it. Other clients: https://adswipefile.com/docs/mcp?utm_source=agent-skill&utm_medium=referral&utm_campaign=get-started. Do not claim the plugin is ready because this skill loaded.
  • For "what can it do", explain without calling a tool.
  • If the user already asked a real question, answer it with the tools instead of stopping at suggesting a question.

What to say

  • AdSwipeFile is an ad library of winning Facebook and Instagram ads: real Meta ads from brands, filed by brand, niche (industry) and format (UGC, before and after, testimonial, comparison…). Ask what a competitor is running, what's working in a niche, or for hooks from real ads.
  • Each ad has its brand, niche, formats, audiences, its copy, an image and a page link. When a list or search comes back cut short (its first few results, with a total), or get_ad returns a breakdown without its layers, the answer carries a note saying so: work from what came, and pass the note and its link on once.
  • Signing in also lets it list the ads the user saved on adswipefile.com, the newest ads of the brands they follow there, and their boards. When the user asks, it can save or unsave an ad and note it, follow or unfollow a brand, and keep boards (create, rename, delete, share, add or remove ads); it changes nothing else.

Creative strategy and the weekly review

For a creative strategy (competitors, personas, concepts, scripts, audits), start with the creative-strategy skill: it routes to the right step. For a recurring read of what followed brands launched, use weekly-competitor-review, by hand or as a scheduled Claude routine or ChatGPT task.

What it cannot do

Say so plainly if asked: it has no spend, reach, CTR or ROAS figures; it cannot create, run or buy ads; it does not generate images or video; it is a curated collection, not a live feed of Meta's ad library; and it changes nothing outside the user's own saves, follows and boards.

Suggested first questions

Offer one, adapted to what the user mentioned:

  • "What ads is Liquid Death running? Sum up its hooks and angles."
  • "Show me winning UGC ads from skincare brands, with their hooks."
  • "Find testimonial ads for supplements and write 3 hooks for my product."
  • "Review what the brands I follow launched this week."

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.