OpenProspect：以一條開源流程取代四款付費銷售工具

I built OpenProspect because the workflow behind a small outreach campaign was scattered across too many tools. Finding a relevant business, identifying a contact, writing a useful email, and tracking the response all lived in separate places.

OpenProspect brings that workflow into one self-hosted project. It combines business discovery, lead enrichment, AI-assisted email drafting, campaign sending, and a dashboard. The goal is to make the process easier to inspect, adapt, and run on your own infrastructure.

How the pipeline works

1. Discover businesses relevant to your campaign.
2. Enrich the records with contact details and context.
3. Draft personalized outreach using AI, then review the copy.
4. Send through your own configured email account.
5. Track campaign activity in the dashboard.

The repository documents a sending cap of 20 emails per hour. That limit matters: it gives you a pace to review outreach quality and avoid treating the tool as a bulk-spam engine. You are responsible for your own contact data, consent, and sending setup.

Why open source and self-hosted?

I wanted a workflow I could understand end to end. With a self-hosted setup, the configuration, prompts, and pipeline are visible. You can change how leads are qualified or how emails are written without waiting for a vendor feature. It also makes it easier to see where the system needs human review.

OpenProspect is a practical starting point for developers and small teams who want to experiment with prospecting and outreach. It is not a promise of automatic sales. Good targeting, a useful offer, and respectful communication still matter more than automation.

The code and setup instructions are here: OpenProspect on GitHub.

I would welcome feedback on the workflow, the developer experience, and the safeguards you would want before using a tool like this.