agentleFS
clairelily2020clairelily2020/workflow-skills/skills/exhibitor-icp-list-builder/SKILL.md

exhibitor-icp-list-builder

Skill1 starsChanged 3 months agoMITView on GitHub

Build verified ICP company lists from trade show exhibitor exports, booth maps, floorplan screenshots, hall maps, logo walls, OCR results, and messy CSV/XLSX files. Use when the user asks to identify exhibitors, extract company names from event images, map a venue zone to companies, clean fake or inactive websites, validate LinkedIn/company pages, deduplicate exhibitor data, score ICP fit, or produce a research-ready B2B lead list. Outputs a tiered company list with source confidence, verification status, reject reasons, fit score, research angle, and recommended next step.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to turn messy event sources into a verified company list.

    line 10
  2. Distinctive:

    Keep aliases in a notes column when OCR or logo matching is uncertain.

    line 30
  3. Distinctive:

    Do not count a company page as active unless there is visible recent activity or enough company evidence to support research.

    line 35
  4. Very specific:

    Never produce a research-ready list without verification_status.

    line 62
  5. Very specific:

    Never let OCR/logo-only matches enter tier A without a second source.

    line 63
  6. Distinctive:

    Never keep fake, dead, or inactive links just to hit a volume target.

    line 64
  7. Distinctive:

    If fewer than 10 usable companies remain, report the shortage and expand sources instead of padding weak rows.

    line 65
  8. Distinctive:

    If an input file has more than 20% rejects, include a short quality note explaining why.

    line 66

≈ “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

Exhibitor ICP List Builder

Use this skill to turn messy event sources into a verified company list. It is for research and qualification only. Do not start outreach, write to a CRM, buy contacts, or launch a campaign unless the user separately approves that step.

Inputs

  • Exhibitor CSV/XLSX exports
  • Floorplan screenshots, hall maps, booth-zone images, or logo walls
  • OCR output from booth maps or images
  • Existing company/contact lists
  • Target ICP definition

Workflow

  1. Separate source types
    • Official exhibitor table: highest-confidence company source.
    • Floorplan/logo/OCR: candidate discovery, not final truth.
    • User-provided list: preserve original rows and add verification columns.
  2. Normalize companies
    • Normalize company name, booth, hall/zone, website, country, category, source file, and source confidence.
    • Deduplicate by website first, then normalized company name, then booth.
    • Keep aliases in a notes column when OCR or logo matching is uncertain.
  3. Verify existence
    • Confirm each company has at least one stable signal: official exhibitor page, working website, active company page, credible marketplace listing, or recent event page.
    • Mark broken websites, fake links, auth-wall-only pages, inactive pages, and unrelated names as reject.
    • Do not count a company page as active unless there is visible recent activity or enough company evidence to support research.
  4. Score ICP fit
    • Score 0-5 for ICP fit, with a short reason.
    • Prioritize companies that sell B2B products, attend events, rely on buyers/distributors, or have a clear sales motion.
    • Down-rank local services, personal brands, one-off retailers, and companies with unclear commercial intent.
  5. Create output tiers
    • A: strong ICP, verified, clear research angle.
    • B: plausible ICP, needs one more verification pass.
    • C: weak ICP or poor contactability.
    • Reject: inactive, fake, broken, duplicate, irrelevant, or cannot verify.

Required Output Columns

company_name, website, linkedin_company_url, country, booth, hall_or_zone, category, icp_segment, icp_score, tier, verification_status, reject_reason, source, source_confidence, research_angle, recommended_next_step, notes

For contact-level lists, add:

contact_name, title, contact_linkedin_url, email_status, page_activity_status, contact_verification_status

Quality Gates

  • Never produce a research-ready list without verification_status.
  • Never let OCR/logo-only matches enter tier A without a second source.
  • Never keep fake, dead, or inactive links just to hit a volume target.
  • If fewer than 10 usable companies remain, report the shortage and expand sources instead of padding weak rows.
  • If an input file has more than 20% rejects, include a short quality note explaining why.

Output Summary

Total input rows:
Verified A:
Verified B:
Rejected:
Top rejection reasons:
Recommended next step:

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.