Skip to content
TakoAPI
Back to skills

data-enricher

by visualdeptcreative

Data & Analytics
1.4kDownloads reported by clawskills.sh
1Stars
1Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Enrich leads with email addresses and format data.

Description

What This Skill Does

Enriches brand leads with contact emails using four fallback discovery methods: website contact pages, Instagram bios, Hunter.io API, and common email pattern guessing. Formats results into structured JSON and syncs them to Notion with deduplication by domain.

Chains four discovery methods in priority order so email lookup succeeds even when a single source returns nothing.

When to Use It

  • Finding contact emails for brand sponsorship outreach
  • Batch processing leads discovered through Meta ads
  • Deduplicating a lead list before importing into Notion
  • Scoring email confidence before sending cold outreach
  • Separating founder personal emails from generic contact addresses

Example Workflow

INPUT

User asks: enrich this lead — domain: brandname.com, ig_handle: @brandname, niche: skincare

AGENT
  1. 1

    Check brandname.com/contact and page footer for mailto: links

  2. 2

    Check @brandname Instagram bio and Contact button for an email address

  3. 3

    Call Hunter.io domain-search API and filter results to confidence score above 70%

  4. 4

    Fall back to pattern guessing (hello@, info@, contact@) if no result found

  5. 5

    Normalize domain, check Notion for an existing entry, then write formatted lead to workspace/leads-enriched-YYYY-MM-DD.json

OUTPUT

{"domain_key": "brandname.com", "brand_name": "Brand Name", "niche": "skincare", "contact_email": "hello@brandname.com", "email_confidence": "high", "email_source": "hunter.io", "status": "new"}

Requirements

  • Hunter.io account and HUNTER_API_KEY environment variable
  • Notion account with a leads database
  • Ollama installed with llama3.2:8b model pulled

Installation

clawhub install visualdeptcreative/data-enricher

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links