agentleFS
prosyo-aiprosyo-ai/gtm-skill/skills/lookalike-account-finder/SKILL.md

lookalike-account-finder

Skill13 starsChanged 25 days agoMITView on GitHub

Use this skill when you want more accounts like your best customers: \"find companies like our top 10\", \"build a lookalike list\", \"who else looks like Acme\". Profiles your best customers on the traits that predicted the win, searches for companies that share them, scores each match on similarity and timing, and explains why each one made the list. Produces a ranked lookalike account list with match reasons and a suggested reference customer for each, ready for outbound or ABM.

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
  1. Distinctive:

    Use this when your best customers are the clearest definition of your ICP.

    line 16
  2. Distinctive:

    Not every shared trait predicts a win.

    line 30
  3. Distinctive:

    MUST explain why each account matched.

    line 70
  4. Distinctive:

    MUST only name a reference customer if the user confirms the customer is approved for that.

    line 71
  5. Distinctive:

    NEVER include current customers, competitors or open deals.

    line 72

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Use this when your best customers are the clearest definition of your ICP. It produces a ranked list of companies that look like them, each with the reason it matched and the customer story to lead with.

Inputs

  • 5 to 20 best customers, ranked by value and fit.
  • Why each one bought, in a line.
  • Exclusions: current customers, open deals, competitors, partners.

The play

1. Profile the seeds

For each seed customer, capture the traits that mattered: industry niche (not just industry), business model, headcount band, growth stage, region, technologies, team structure, the trigger that preceded the deal. Look for the 3-5 traits most seeds share; ignore traits only one or two have.

2. Weight the traits

Not every shared trait predicts a win. Weight by how specific it is: "sells to restaurants through a field sales team" matters more than "B2B, 50-200 employees".

3. Search in widening circles

  1. Direct peers: competitors of your customers, members of the same associations, exhibitors at the same niche events.
  2. Same model, adjacent niche: a company selling the same way to a neighbouring market.
  3. Same stack and stage: companies running the same tools at the same size.

Direct peers convert best because the customer story transfers word for word.

4. Score each match

ComponentPoints
Shares the core niche and model0-40
Size and stage within seed range0-20
Shares stack or team structure0-15
Active trigger in last 90 days0-25

70+ goes to outbound now; 50-69 goes to a nurture or a later wave.

5. Attach the reference

For each account, name the seed customer it most resembles. That becomes the proof point in the first message, if the customer is approved for reference.

Output format

Seeds: <list>  |  Shared traits: <3-5 weighted traits>
Lookalikes (ranked):
  <company> - score - match reasons (2-3) - trigger - closest seed - approved to name? y/n
Excluded: <counts by reason>

What good looks like

  • Match reasons a rep could say out loud: "Same model as Northwind: field teams selling into independent pharmacies."
  • Direct peers of happy customers lead the list.
  • Timing is scored, so the top of the list is not just similar but ready.
  • The common mistake is matching on industry and size only, which returns hundreds of generic companies and none of the nuance that made the seeds buy.

Rules

  • MUST explain why each account matched.
  • MUST only name a reference customer if the user confirms the customer is approved for that.
  • NEVER include current customers, competitors or open deals.

Example prompts

  • "Here are our 12 best customers. Find 100 lookalikes and rank them."
  • "Find companies like Northwind and Contoso and tell me which customer story to use for each."
  • "Build a lookalike list from our top accounts in healthcare only."

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.