Verified project record
OrchestratorInc/agent-orchestrator
A desktop workspace that plans, runs, and supervises multiple coding agents with isolated workspaces and a live Kanban board tracking pull requests, CI, and reviews. It targets coordination problems that arise when running several agents on one project.
Project overview
It derives board-card position from session, pull request, CI, and review facts, and gives each worker its own branch and worktree to prevent collisions when parallel agents edit the same repository.
- Project type
- AI Agent · AI Coding
- Use cases
- Coding & Development
- Deployment
- Refer to project documentation
- License
- Apache-2.0
Best for
- Developers who run multiple coding agents on a single Git repository and need help splitting work, isolating branches, and tracking changes through review and merge.
- Users who want to supervise agent sessions through a desktop GUI rather than scripts or CLI tooling.
- Teams or individuals already using coding agent harnesses such as Claude Code or Codex who want them coordinated in one workflow.
Key capabilities
- A persistent planning agent that breaks outcomes into focused tasks, spawns or redirects workers with relevant context, and coordinates follow-up work.
- A live board deriving card position from session, pull request, CI, and review facts, showing working, blocked, in-review, and ready-to-merge states.
- Each worker gets one task, one coding agent, and one isolated workspace with its own branch and worktree for Git-backed work.
Limitations and risks
- Worker sessions need a user-installed coding agent harness such as Claude Code or Codex, and a model must be chosen when creating each worker session; models are supplied by the selected harness, not by this application. Costs, if any, depend on that harness and model.
- No end-to-end determination of a local-only, offline, or private data boundary is documented. Local orchestration of the desktop app does not by itself establish where inference happens, since models come from the selected harness.
- GPU requirements, minimum hardware, and cost dependencies are not documented; verify these against your chosen harness and model before adopting.
- The documented interaction mode is a desktop GUI; whether other interaction modes, such as web or headless access, are available is not documented here, so verify your access requirements before relying on them.
Getting started
- Setup is documented as easy: download the desktop app for your platform, open it and point it at the repository to manage, then create a worker with New task. Inputs needed are a Git repository and task descriptions, with a chosen coding agent and model.
Alternatives and comparisons
- Instead of orchestrating multiple agents in parallel, it provides a software development methodology for AI coding agents that enforces test-driven development and subagent-driven development, addressing agents that skip planning and testing.
- Provides small, composable engineering skills enforcing fundamentals like test-driven development and code review, targeting code quality and intent alignment rather than multi-agent coordination.
- Provides a large set of reusable skills, rules, and hook scripts across many harnesses including Claude Code and Cursor, complementing harness workflows rather than supervising parallel workers on a board.
Project comparisons
Evidence and sources
- GitHub project description: Run and supervise teams of coding agents from planning to merge. Any harness (Claude code, codex, +25 more). Desktop, web, mobile, and cloud agents.
- README: One coding agent can handle a task. Running several across a project creates a different job: deciding what matters, splitting work cleanly, giving each agent the right context, p…
- README: Orchestrator.inc is a local desktop workspace built for that job. Add a repository and create a worker session with the coding agent, model, and interface that fit the task. For G…
- README: Use **New task** when the work is already clear. Describe the outcome, choose an agent and model, attach relevant files, and work with the agent in structured Chat or its native t…
- README: When a plan becomes actionable, the orchestrator can break it into focused tasks, spawn or redirect workers, pass each worker the relevant context, follow their progress, and coor…
Search
Find projects, verify facts, compare options, and filter by your requirements
Try a searchA click only fills the search box; you stay in control
Project Details
0