Automated AI Lead Generation with Open Source Worker

Description and use

The author describes using an open-source AI tool, 'Lead Researcher', to generate 2,000 qualified leads over a weekend with zero tool costs. The system runs locally via AI coding tools (Claude Code, Codex CLI), researches public data, verifies ICP fit, and exports a CSV with evidence.

Problem

Traditional lead generation requires significant human effort, expensive tools, or subscriptions to research companies, identify decision-makers, verify data, and score leads against an Ideal Customer Profile (ICP).

Outcome

Generated 2,000 qualified leads over a weekend with zero spend on lead gen tools. The system automatically researched, verified, and scored leads based on a strict ICP, exporting a CSV with evidence for each lead.

Workflow

• Install the 'Lead Researcher' worker into an existing AI coding tool (Claude Code, Codex CLI, or OpenCode).
• Define the Ideal Customer Profile (ICP) in a plain-language file.
• Issue a command to the worker, e.g., 'Get 2000 leads for me based on my ICP and don't stop until it is done.'
• The worker researches public web data, identifies people, verifies data, and scores leads against the ICP.
• The worker rejects duplicates, guessed emails, and vague fits based on strict quality gates.
• The worker exports a 'leads.csv' file containing the lead list, sources, and justifications.
• Optionally, use the hands-free scheduler to run the worker periodically in the background.

Evidence

My AI company generated 2,000 qualified leads over the weekend, while I slept. Zero dollars spent on lead gen tools.

Source

A lead only counts if it scores above the bar AND cites the ICP file as evidence.

Source