agentleFS
kelpi-aikelpi-ai/google-ads-skills/skills/responsive-search-ad-writer/SKILL.md

responsive-search-ad-writer

Skill2 starsChanged 4 months agoMITView on GitHub

Write a complete Google responsive search ad from real query intent and supported page claims, then validate every asset and character limit. Use when an ad group needs credible copy, sitelinks, and callouts.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When a focused Search ad group needs a new or replacement responsive search ad.

    line 14
  2. Distinctive:

    Use a connected Google Ads MCP for read-only context, or paste the keywords, search terms, current ads, page copy, final URL, brand voice, …

    line 15
  3. Distinctive:

    Never add "no credit card", "guaranteed", setup-time, savings, security, certification, or performance claims without supplied proof.

    line 46
  4. Distinctive:

    Do not invent sitelink destinations.

    line 47

≈ “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

Responsive Search Ad Writer

Doctrine

A responsive search ad is a set of assets Google can combine in many orders. Every headline and description must work alone and beside the others. Start from the query and the page, use only claims the business can prove, and validate the copy before calling it ready.

When to use

  • When a focused Search ad group needs a new or replacement responsive search ad.
  • Use a connected Google Ads MCP for read-only context, or paste the keywords, search terms, current ads, page copy, final URL, brand voice, and policy constraints.
  • Supply real sitelink URLs and supported facts. Missing proof stays missing; it is never filled with a plausible claim.

Run it

Draft a Google responsive search ad. Use live account data for read-only context if connected; otherwise use what I paste. Do not create or edit an ad.

1. Confirm the ad-group job, target queries and keywords, final URL, page facts, brand voice, legal requirements, current assets, and approved sitelink URLs.
2. Make a claim ledger before writing: supported facts, unsupported claims to avoid, and missing proof.
3. Output in this order:
   - ad-group and query map;
   - negative-keyword questions, not automatic negatives;
   - 4 or more sitelinks with supplied URLs, or clearly marked URL gaps;
   - 4 or more callouts;
   - final URL and 2 display paths;
   - 15 distinct headlines;
   - 4 distinct descriptions;
   - pinning plan, default unpinned unless a legal or message requirement must always show;
   - validation report and claim sources.
4. Enforce current RSA limits: every headline at most 30 characters, every description at most 90, and every path at most 15. Print the character count beside each asset.
5. Make assets useful in different combinations. Avoid repeating the same claim with tiny wording changes.
6. Map every factual claim to supplied page evidence. If a sitelink URL, price, trial condition, certification claim, timing claim, or feature is not supplied, do not invent it.
7. Check that the first 3 headlines can appear together, that any headline can pair with any description, and that no combination overclaims.

Return copy and a pass/fail validation report. If required inputs are missing, label the draft "not ready to publish" and list the gaps. End with "No ads were created or changed."

Guardrails

  • Google allows up to 15 headlines and 4 descriptions. This skill writes the full set for range, but does not pretend every asset will serve.
  • Never add "no credit card", "guaranteed", setup-time, savings, security, certification, or performance claims without supplied proof.
  • Do not invent sitelink destinations. A useful label with no verified URL is a gap, not a ready asset.
  • Pinning reduces combination flexibility. Use it only when the user supplies a real requirement.
  • Copy is a draft. Creating, editing, or enabling an ad requires a separate, explicit approval step.

Good output looks like

One intent-matched RSA with validated counts, distinct assets, a visible claim ledger, honest URL gaps, and no unsupported promise hiding in the copy.

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.