wdm0006/evergreen-skills/contact-capture

Smart Contact Capture

Works with Evergreen, a local-first personal CRM for macOS. Get it on the Mac App Store.

When to Use

• Someone shares a business card or email signature
• You have raw text with contact details (from a conference, email thread, or LinkedIn)
• You need to batch-create contacts from meeting notes or an event attendee list
• You received an introduction email and want to add the new person

How It Works

1. Parse the unstructured text to extract: name, title, organization, email, phone, location, and any contextual notes
2. Search existing contacts with search_contacts to check for duplicates
3. If the contact doesn't exist, create it with create_contact
4. Add relevant tags via the contact's tags field with update_contact (e.g., "met-at-conference", "investor", "lead")
5. Record context in the contact's notes field with update_contact (how you met, who introduced you, conversation topics)
6. If someone introduced you to this new contact, resolve that introducer to an existing contact with search_contacts. If they're already in the CRM, call record_introduction(introducerId, newContactId, notes?) to record the introducedBy edge (who brought this person into your network). Keep the free-text note from step 5 regardless — the edge complements it. If the introducer isn't a known contact, skip the call and rely on the note alone.
7. If an introduction, log the interaction with log_interaction

Extraction Patterns

Example

Input:

Hey, great meeting Sarah Chen at the AI dinner last night. She's the
CTO at Meridian Health, based in Atlanta. Her email is sarah@meridianhealth.com
and she's really interested in our API work. David Kim introduced us.

Actions:

1. search_contacts({ query: "Sarah Chen" }) → no match
2. create_contact({
     name: "Sarah Chen",
     title: "CTO",
     organization: "Meridian Health",
     emails: ["sarah@meridianhealth.com"],
     location: "Atlanta"
   }) → sarah_id
3. update_contact({ contactId: sarah_id, tags: ["ai", "healthcare", "atlanta"] })
4. update_contact({ contactId: sarah_id, notes: "Met at Atlanta AI Dinner. Introduced by David Kim. Interested in our API work." })
5. search_contacts({ query: "David Kim" }) → matches contact david_id
6. record_introduction({ introducerId: david_id, newContactId: sarah_id, notes: "Introduced Sarah at the Atlanta AI Dinner" })
7. log_interaction({ contactId: sarah_id, type: "meeting", description: "Met at Atlanta AI Dinner — discussed API work" })

(If search_contacts finds no match for the introducer, skip step 6 and keep the note from step 4.)

Checklist

Contact Capture:
- [ ] Full name captured in the single `name` field (no first/last split)
- [ ] Duplicate check performed before creating
- [ ] All available fields populated
- [ ] Tags applied for context/source
- [ ] Meeting context saved in notes
- [ ] Introducer edge recorded with `record_introduction` when the intro's source is a known contact
- [ ] Initial interaction logged

Learn More

• Evergreen — Local-First Personal CRM
• Vibe Coding a Personal CRM
• Evergreen Gets Serious
• Evergreen Gets Even Evergreener