linkedin-skills

name	linkedin-skills
description	Three LinkedIn automation skills — (1) auto-apply to Easy Apply jobs, (2) scrape profile data by company/country/industry, (3) discover contacts via BFS/DFS for referrals/networking with email generation. Uses Playwright MCP browser automation.

LinkedIn Skills

Three skills for LinkedIn automation via Playwright MCP browser tools.

---

Skill 1: Job Auto-Apply

Automatically apply to LinkedIn Easy Apply jobs with target-based stopping and keyboard controls.

Usage: /linkedin-job-auto-apply

Full docs: skills/linkedin-job-auto-apply/SKILL.md

// Paste autoApplyLinkedInJobs.js, then:
await autoApplyLinkedInJobs(page, {
  targetApplications: 20,
  searchKeywords: 'software engineer',
  location: 'United States'
});
// P=Pause  R=Resume  Q=Quit

---

Skill 2: Profile Scraper

Search LinkedIn people by company, country, and industry — scrape name, current company, location, work history, and industry for each.

Usage: /linkedin-profile-scraper

Full docs: skills/linkedin-profile-scraper/SKILL.md

// Paste scrapeLinkedInProfiles.js, then:
const results = await scrapeLinkedInProfiles(page, {
  company: 'Google',
  country: 'United States',
  industry: 'Software Development',
  maxProfiles: 20
});
console.log(JSON.stringify(results, null, 2));

---

Skill 3: Contact Reacher

Systematically discover LinkedIn contacts for job referrals or networking using BFS/DFS traversal. Extracts LinkedIn URLs, generates company email candidates, optionally sends connection requests, and saves output locally as JSON + CSV.

Usage: /linkedin-contact-reacher

Full docs: skills/linkedin-contact-reacher/SKILL.md

// Paste discoverContacts.js, then:
const contacts = await discoverContacts(page, {
  seeds: [{ type: 'search', company: 'Google', role: 'Software Engineer' }],
  strategy: 'bfs',
  maxContacts: 30,
  maxDepth: 1,
  targetCompanies: ['Google'],
});

// Paste extractContactInfo.js, then:
const enriched = await extractContactInfo(page, contacts, {
  companyDomains: { 'Google': 'google.com' },
});

// Paste saveOutput.js, then:
await saveOutput(enriched, { label: 'google-referrals', format: 'both' });
// Output: ./output/google-referrals_2026-04-27T....{json,csv}

---

Specialized Agents

All three skills are backed by seven specialist agents for deeper control:

Agent	Skill Name	Purpose
Strategy Agent	linkedin-strategy-agent	Filter & score jobs, blocklist, seniority matching, session budget
Automation Agent	linkedin-automation-agent	Timing, retry logic, rate limiting, anti-detection patterns
Web Structure Agent	linkedin-web-structure-agent	LinkedIn DOM selectors, lazy loading, resilient element targeting
QA Agent	linkedin-qa-agent	Pre-flight checks, result verification, data quality reports
Contact Discovery Agent	linkedin-contact-discovery-agent	BFS/DFS traversal strategy, seed selection, depth tuning
Outreach Agent	linkedin-outreach-agent	Connection note templates, rate limits, acceptance rate optimization
Email Generator Agent	linkedin-email-generator-agent	Email pattern generation, domain inference, confidence scoring

Full agent docs: skills/agents/

Orchestrated run flow:

1. QA Agent       → preFlightCheck(page)        # must PASS — abort if fails
2. Strategy Agent → filterJobs(jobs, prefs)      # score & filter before applying
3. [run skill with filtered list]
4. QA Agent       → verify / generateReport()   # after finishing
   ↓ selectors broken    → Web Structure Agent
   ↓ high failure rate   → Automation Agent
   ↓ irrelevant matches  → Strategy Agent

---

Installation

# Claude Code marketplace
/plugin marketplace add yennanliu/linkedin-skill
/plugin install linkedin-job-auto-apply
/plugin install linkedin-profile-scraper

# Or via install script (auto-detects Claude/Gemini)
./install.sh

See INSTALLATION.md | INSTALL_GEMINI.md | INSTALL_COPILOT.md