CRM

Comp AI CRM is an open source customer relationship management tool built around the idea that sales teams shouldn't be manually updating records. Instead of requiring reps to log every interaction, agents do it for them: reading the team's inbox, pulling in new contacts from email threads, enriching company profiles with industry data and logos, and scheduling their own follow-ups.

The core idea is that agents don't just automate tasks, they also create other agents. Describe a workflow in plain language and the system generates a new agent to handle it on its own schedule. Out of the box, suggested workflows include things like briefing a deal owner before a renewal call, flagging deals with no activity in 14 days, or handing new customers from sales to onboarding.

Key capabilities:

• Inbox-driven record creation: A new person on an email thread becomes a contact automatically, with their company record populated alongside it.
• Company and contact enrichment: Profiles fill in with domain, industry, logo, and last activity without manual input.
• Self-scheduled follow-ups: Agents book their own rechecks and can explain why they'll return at a specific time, not just when.
• Agent builder: Describe a process in a sentence and the system writes and queues a new agent to run it.
• Per-record Q&A: Ask any company or contact a question and the agent reads their site plus your interaction history to answer, showing its reasoning.
• Slack integration: Agents can ping team members or create Slack channels based on deal events like closed-won or overdue contracts.

Built with Next.js and deployable via Vercel, it's designed to be forked and customized. The pipeline view, company table, deal tracking, and activity feed are all standard CRM surfaces, but the agent layer sits on top of all of them.