Comp AI CRM is a customer relationship management system designed for AI agents to operate autonomously rather than as a supplementary feature.
The tool addresses the problem that traditional CRMs require humans to manually research and enter customer data, while AI-augmented CRMs still treat the agent as a peripheral feature. This system inverts that relationship: the agent runs on its own deployment and schedule, maintaining its own work queue, deciding what to investigate next, and booking follow-ups independently. The agent operates continuously even when the browser is closed. The core principle is that nothing about a person is guessed—tools report only what they observed, and a ledger system weights evidence strength. Strong evidence updates customer records directly, while weak evidence surfaces as suggestions for human review, preventing confidently incorrect data from corrupting the database.
Teams should adopt this tool if they want customer research and data enrichment to happen continuously in the background without human intervention for each task. It suits organizations that need systematic, evidence-based customer intelligence rather than manual data entry or confidence-scored guesses. The project is built on eve, Vercel's filesystem-first framework for durable agents, where tools, skills, and schedules are defined as files and the runtime handles session persistence across redeployments.
The project maintains active development with regular commits addressing core functionality and agent capabilities. The codebase shows ongoing refinement of the agent's decision-making logic and the evidence-weighting system that determines what gets written to customer records. Work continues on expanding the tool ecosystem that agents can use to gather customer information.
