ChatML is a desktop tool for orchestrating Claude Code sessions. It is built around parallel sessions, code review, cost tracking, and pull request workflows in one interface, with an emphasis on keeping agent activity visible while work is in progress.
The app lets a user scope tasks and delegate them to agents in isolated Git worktrees, so multiple features can be handled at the same time without branch conflicts or constant context switching. It includes a dashboard for seeing all running sessions at a glance, showing status, progress, and what each agent is working on. ChatML also provides a built-in diff viewer with inline comments, and it supports Quick, Deep, and Security review modes before changes are merged. Real-time cost and token tracking is available per session, with budget limits. Other documented features include an integrated terminal, a command menu with keyboard shortcuts, multi-workspace support, a file browser with tabs, activity notifications, and support for MCP servers. It also includes Linear integration, a skill library, plan mode with approval, sub-agent spawning, and extended thinking.
ChatML is described as a native desktop application for macOS, Windows, and Linux. The interface and implementation stack are listed as Tauri for desktop, Rust for the backend, Golang for the API, and Next.js for the UI. The project is 100% open source and uses the GPL-3.0 license. The site also says it is built in public and backed by the community.
In the Coding AI & assistants space, ChatML takes a focused approach. It focuses on coordinating multiple AI coding agents to accelerate software development and code review. ChatML is an open-source project aimed at software developers. ChatML is open source under the Open Source license. It runs on the command line, macOS, Windows, and Linux.
ChatML first shipped in 2026. The project is developed in the open on GitHub with 46 stars and 88 commits in the last 90 days. Key capabilities include parallel agent orchestration, code review, and cost tracking.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do