linkedin-recon

star 6

Browser-based LinkedIn profile recon using Browserbase MCP. Extracts recent posts, activity, and personalization hooks for outreach. Use when the user types /linkedin-recon.

shawnla90 By shawnla90 schedule Updated 3/9/2026

name: linkedin-recon description: Browser-based LinkedIn profile recon using Browserbase MCP. Extracts recent posts, activity, and personalization hooks for outreach. Use when the user types /linkedin-recon.

LinkedIn Recon

Commands

  • /linkedin-recon or /linkedin-recon accepted - recon accepted connections from latest HeyReach export CSV
  • /linkedin-recon <url> - recon single profile
  • /linkedin-recon <csv-path> - recon from custom CSV

Workflow

  1. Identify profiles: parse CSV or URL, deduplicate
  2. Create Browserbase session: browserbase_session_create
  3. Visit each profile (5-10s between visits, max 15 per session):
    • Activity page: {url}/recent-activity/all/ - extract posts, dates, engagement, hashtags
    • Profile page: {url} - extract headline, about, experience, skills
  4. Compile recon: per contact - about, recent activity, personalization hooks (3), suggested message
  5. Save: individual files to resources/recon/{username}-recon-{date}.md + summary CSV
  6. Close session: browserbase_session_close

Rate Limiting

  • 5-10s between profiles, max 15 per session
  • READ ONLY - never connect/like/message
  • Stop immediately on CAPTCHA/block

Output

Summary table: profiles scanned, active vs inactive, per-contact hooks. Individual recon markdown files.

Install via CLI
npx skills add https://github.com/shawnla90/shawn-gtme-os --skill linkedin-recon
Repository Details
star Stars 6
call_split Forks 2
navigation Branch main
article Path SKILL.md
More from Creator