ProwlA 🐾

An agentic job hunting assistant. Track applications, research companies, and generate personalized materials — manually or with AI agents.

Why ProwlA?

Job hunting is tedious. You're juggling spreadsheets, researching companies, writing cover letters, and tracking where you applied. ProwlA automates the boring parts so you can focus on what matters — landing the right role.

What makes it different:

• 🤖 Agent-first architecture — 28 MCP tools let AI agents help with research and materials
• 🔍 Deep company research — Funding, team, contacts, fit analysis
• ✉️ Personalized materials — Cover letters and outreach emails tailored to each role
• 📊 Smart tracking — Priority scoring, status management, follow-up reminders
• 🎯 Configurable criteria — Define your ideal job, let ProwlA find matches

Quick Start

Prerequisites

• Node.js 18+
• npm or yarn

Installation

# Clone the repository
git clone https://github.com/yourusername/prowla.git
cd prowla

# Install all dependencies
npm run setup

# Copy example configs
cp config/profile.example.json config/profile.json
cp config/search.example.json config/search.json
cp .env.example .env

# Add your Anthropic API key to .env (optional, for AI features)
echo "ANTHROPIC_API_KEY=your-key-here" >> .env

# Start the app
npm run dev

Open http://localhost:3001 in your browser.

Modes of Operation

ProwlA works in three modes depending on your needs:

🖱️ Manual Mode

Just use the web UI. Add jobs manually, update statuses, take notes. No AI required.

Best for: People who want a clean ProwlA without automation.

⚡ MCP Assisted Mode

Connect ProwlA to Cursor or Claude Desktop via MCP. Ask the AI to research companies, generate materials, or find new opportunities.

Best for: Developers using AI-assisted coding tools who want on-demand help.

You: "Research the company Acme AI and generate a cover letter"
Agent: *uses ProwlA MCP tools to research and save materials*

🤖 Autonomous Mode

Set up scheduled scans and heartbeat processing with OpenClaw. ProwlA runs in the background, finding opportunities and researching them automatically.

Best for: Power users who want a fully automated job search pipeline.

See docs/modes/ for detailed setup instructions.

Features

Job Tracking

• Add jobs from any source (job boards, referrals, cold outreach)
• Track status: Not Started → Researching → Applied → Interviewing → Offer
• Priority levels with visual indicators
• Filter and search across all applications
• "NEW" badges for recent additions

Company Research

• AI-powered deep research (funding, team, product, culture)
• Contact discovery (LinkedIn, email)
• Fit analysis based on your profile
• Structured notes following a consistent template

Application Materials

• Personalized cover letters
• Tailored outreach emails
• Generated from your profile + company research

MCP Integration

28 tools across 8 categories:

Category	Tools	Description
Jobs	7	CRUD operations, bulk updates, fuzzy search
Research	4	Company research notes management
Materials	3	Cover letters and outreach emails
Tasks	7	Task queue with locking mechanism
Events	4	Event subscription and polling
Query	4	Specialized queries and statistics
Config	3	Settings and preferences
Tokens	3	Usage tracking and cost estimates

See mcp-server/README.md for full tool documentation.

Configuration

ProwlA is fully configurable for your job search:

config/profile.json — Your Background

{
  "name": "Your Name",
  "title": "Product Designer",
  "background": "10 years of experience in...",
  "skills": ["UX Research", "Prototyping", "Design Systems"],
  "highlights": ["Led redesign that increased conversion 40%"]
}

config/search.json — Job Criteria

{
  "targetRoles": ["Product Designer", "UX Lead", "Design Director"],
  "workTypes": ["Full-time", "Contract", "Fractional"],
  "remote": true,
  "compensation": {
    "salary": { "min": 150000 },
    "hourly": { "min": 100 }
  }
}

See docs/CONFIGURATION.md for all options.

Architecture

prowla/
├── client/          # React UI (Vite)
├── server/          # Express API + SQLite
├── mcp-server/      # MCP tools for AI agents
├── config/          # User configuration (profile, search, sources)
├── scripts/         # Job scanning scripts (user-configurable)
├── tasks/           # Task queue (for async processing)
└── docs/            # Documentation

Tech Stack

• Frontend: React 18 + Vite
• Backend: Express.js + better-sqlite3
• Database: SQLite with WAL mode
• MCP Server: @modelcontextprotocol/sdk
• AI: Anthropic Claude API (optional)

Documentation

Document	Description
INSTALLATION.md	Detailed setup guide
CONFIGURATION.md	All configuration options
JOB-SOURCES.md	Configure job board sources
modes/MANUAL-MODE.md	Using without AI
modes/MCP-MODE.md	Cursor/Claude Desktop integration
modes/AUTONOMOUS-MODE.md	OpenClaw setup
agents/MCP-REFERENCE.md	All 28 MCP tools
agents/WORKFLOW-EXAMPLES.md	Agent workflow patterns

Contributing

Contributions are welcome! See CONTRIBUTING.md for guidelines.

Development

# Start in development mode
npm run dev

# Run client only
npm run client

# Run server only
npm run server

# Run MCP server (for testing)
node mcp-server/index.js --list

License

MIT — see LICENSE for details.

Author

Stu Green — AILOGY Labs

Built with the help of Claude by Anthropic.

Acknowledgments

• Anthropic — Claude AI and MCP SDK
• Inspired by the pain of job hunting

---

ProwlA — Stop hunting. Start prowling. 🐾

Made with 🐾 by AILOGY Labs