Monitor AI agent sessions locally with this open-source dashboard

Running several AI coding assistants in parallel can quickly become unmanageable. Between Claude Code sessions, dynamically spawned sub-agents, and tools like OpenCode running simultaneously, developers often struggle to maintain visibility into their workflows. Without a unified view, tracking active conversations, pending approvals, and input requirements becomes a manual and error-prone process.

To address this, a developer has released CrewView—a lightweight, open-source dashboard designed to bring clarity to AI agent operations. The tool provides a real-time, local interface for monitoring every AI coding session running on your machine, eliminating the need for cloud services or user accounts.

Track sessions, status, and dependencies in one place

CrewView aggregates all active AI agent sessions in a centralized list, displaying key details such as conversation status, elapsed time, and a preview of the current prompt. The dashboard organizes sessions into distinct workflow stages using a Kanban-style board, categorizing them as Running, Needs Approval, or Needs Input. This structured layout helps developers prioritize tasks and respond to agent requests without losing context.

For those using agent frameworks that spawn intermediate or child processes—such as Claude Code’s Task agents—CrewView visually nests sub-agent sessions under their parent, maintaining a clear hierarchy. This feature ensures that even nested workflows remain comprehensible, reducing the risk of oversight in complex operations.

Build and schedule multi-step AI agent pipelines

Beyond real-time monitoring, CrewView includes a workflow builder that lets developers define multi-step agent pipelines. These pipelines can be triggered manually or scheduled to run at specific times, enabling automated execution of routine tasks. The tool supports multiple agent sources, including Claude Code—via local JSONL files—and OpenCode, which interacts through an HTTP API. This flexibility allows teams to integrate CrewView into existing workflows without vendor lock-in.

All data captured by CrewView is stored locally in a SQLite database, ensuring privacy and eliminating reliance on external servers. The application ships as a single Go binary with an embedded React dashboard, requiring no Docker setup, cloud accounts, or telemetry beyond user consent. Installation is straightforward, with a one-command script available from the project’s GitHub repository.

Open-source, MIT-licensed, and built for developer productivity

CrewView is released under the MIT license, encouraging community contributions and customization. The project’s GitHub repository serves as the central hub for documentation, issue tracking, and feature requests. Developers using AI coding agents are encouraged to share their feedback or suggest new visibility features that could improve their workflow.

For those managing complex AI-driven development environments, CrewView offers a practical solution to regain control over agent sessions. By centralizing monitoring and workflow management in a local-first tool, it reduces cognitive load and enhances productivity without introducing unnecessary complexity.