LinkedIn Research

Scan LinkedIn profiles for VC signals relevant to KTLYST fundraising.

Steps

1. •

   Read scripts/social_scan_manifest.json and extract metadata.linkedin_scan.last_run

2. •

   Report to user:

   • •Last LinkedIn scan date/time
   • •How long ago that was
   • •Number of Priority 1 profiles to scan
3. •

   Ask: "Ready to watch the browser? (y/n)"

4. •

   Wait for user confirmation before proceeding

5. •

   If confirmed, use browser automation to:

   • •Navigate to each Priority 1 LinkedIn profile from the manifest
   • •Check recent activity (last 14 days only)
   • •Look for investment signals, thesis statements, portfolio news
   • •When creating signals, extract the specific post URL (not profile URL)
   • •Create signal entries in signals-data.json
6. •

   After completion:

   • •Update metadata.linkedin_scan.last_run in scripts/social_scan_manifest.json to current timestamp
   • •Update metadata.linkedin_scan.profiles_scanned count
   • •Commit and push changes
   • •Report summary of signals found

Important Rules

• •Never run LinkedIn automation without explicit user confirmation
• •User must be watching to handle CAPTCHAs or warnings
• •Extract specific post URLs, not profile URLs
• •Only look at posts from the last 14 days