search-intent-mapper
Skill2 starsChanged 4 months agoMITView on GitHub
Group Google Ads queries by the answer each searcher needs, then map every cluster to one ad-group promise and a page that can support it. Use before adding keywords or restructuring Search.
In this repository11 other files it gives its agents.
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
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
- Distinctive: line 14
Before building or restructuring Search campaigns and ad groups.
- Distinctive: line 15
Use a connected Google Ads MCP for read-only context, or paste terms, keywords, current structure, ads, final URLs, and page copy.
- Distinctive: line 37
Do not mix competitor, brand, pricing, and category terms just because they share a product word.
- Distinctive: line 38
Do not send every cluster to the homepage.
- Distinctive: line 39
Keep informational traffic out of a direct-response campaign unless the user has an explicit content objective and a fitting page.
≈ “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.
Files
Search Intent Mapper
Doctrine
Keywords that share words do not always share a job. Pricing, comparison, login, category shopping, and education each need a different answer. Build clusters from the query language, then give each cluster one promise and one credible destination. If the site cannot answer an intent, call it a gap instead of routing everything to the homepage.
When to use
- Before building or restructuring Search campaigns and ad groups.
- Use a connected Google Ads MCP for read-only context, or paste terms, keywords, current structure, ads, final URLs, and page copy.
- Bring the offer, supported audiences, brand terms, competitors, and business exclusions.
Run it
Build a READ-ONLY Google Search intent map. Use live account data for context if connected; otherwise use what I paste. Change nothing.
1. Confirm the offer, valid audiences, current campaigns and ad groups, available pages, and business exclusions.
2. Read each query as a job. Separate navigation, pricing, category shopping, comparison, competitor, audience-specific, problem research, and other distinct intents only when the words support that distinction.
3. For each cluster show: intent label, representative queries, the searcher's question, proposed campaign and ad group, one message promise, best final URL, and evidence gaps.
4. Keep each proposed ad group focused on one answerable job. If two terms need different promises or pages, split them.
5. Route brand login to the login page and brand pricing to pricing when those pages exist. Do not send navigation to a generic product page.
6. Put unsupported educational or template intent in a no-fit or content-gap list instead of buying it by default.
7. Flag claims the current page cannot support and name the page or copy needed before launch.
Return the intent map, no-fit terms, page gaps, and a draft structure. Mark every structural change "approval required". End with "No changes were made."Guardrails
- Derive intent from query language and supplied context, not a generic funnel label.
- Do not mix competitor, brand, pricing, and category terms just because they share a product word.
- Do not send every cluster to the homepage. A page must answer the query and support the ad promise.
- Keep informational traffic out of a direct-response campaign unless the user has an explicit content objective and a fitting page.
- This is a structure draft. It never creates campaigns, ad groups, keywords, ads, or negatives.
Good output looks like
A clean map where every query has one job, every ad group gives one answer, every page supports the promise, and unsupported intent is called out instead of forced into the account.
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
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.

