Background execution with review gates
Convert prompts into task cards, start work by moving cards into progress, and keep final responses, transcripts, and timelines attached to the task.
Vibecoding command center for solo builders
Run Claude Code, OpenAI Codex, and built-in coding agents from one workspace with Kanban execution, swarm coordination, mux terminals, reusable Skill management, ten visual themes, and readable tool activity that keeps you in creative control.
Product
Move between four focused views for delegated tasks, multi-agent coordination, terminal-heavy sessions, and reusable agent skills without leaving your local workspace.
Convert prompts into task cards, start work by moving cards into progress, and keep final responses, transcripts, and timelines attached to the task.
Build graph-based agent setups with role templates, connected messaging, shared workspaces, and provider-specific terminal choices.
Manage template-driven sessions for repeated workflows where you need dense terminal control and fast switching across agent contexts.
Scan, preview, import, and install SKILL.md capabilities across central storage, supported agent platforms, and project-local discovery.
Themes
Change the entire application surface without changing the workflow. Choose from five focused dark themes and five expressive light themes, each applied consistently across Kanban, Swarm, Mux, and Skill modes.
High-contrast green phosphor styling for terminal-first focus.
Quiet panels and softened geometry for long building sessions.
Bright rounded surfaces with an accessible purple hierarchy.
Presentation-ready black and gold with crisp geometric framing.
Focused contrast for terminals, graphs, and long-running work.
Clear daylight surfaces ranging from strict to approachable.
Tool Transparency
The transparency feature turns agent internals into concise lifecycle events: queued, running, completed, and failed. Users see what tool is being called, why it is being used, and the high-signal result without raw JSON noise.
Open source files needed to inspect Kanban timeline rendering.
Find references to built-in agent tool event parsing.
Added bounded previews for tool summary and result details.
Command failed with permission error. User approval required.
Users saw coarse status changes and generic queued signals. Reviewers had to infer progress from raw terminal output and final replies.
Each tool call appears as a stable row with action summary, lifecycle state, concise result preview, and visible failure context.
Transparency reduces guesswork, keeps long-running work understandable, and makes it easier to trust an agent while staying in the loop.
Why It Works
Dorchestrator is not another model wrapper. You bring your own provider keys and get a local control layer for running agents, reusing workflows, and understanding tool activity while the code changes.
Solo developers, indie hackers, technical founders, and AI-heavy builders who want several coding agents working from one local workspace.
Agent output is useful, but raw terminals become hard to supervise once multiple long-running tasks are happening at the same time.
A better local workflow: reusable templates, scheduled tasks, run history, readable tool activity, and less context switching between terminals.
Choose Pro for continuous updates or Lifetime for a durable desktop license. Models stay on your own provider accounts, so the app never resells inference.
Practical demos, real screenshots, short workflow videos, and examples that show how one builder can keep several agents moving without chaos.
Workflows, templates, run history, and local workspace context become a personal operating memory for the way you build software with agents.
Early Access
Dorchestrator is still being built. Join the waitlist for launch updates, early-access invitations, and a clear notice when paid plans become available.
Plans
Choose a subscription for continuous updates, or own the desktop license with a one-time purchase. Your model usage stays on your provider accounts.
For daily builders who want every workspace mode and continuous product updates.
For builders who prefer one payment and a desktop version they can keep.
Workflow
Start from the tools you already use: Claude Code, Codex, or the built-in coding agent, all pointed at the same local workspace.
Convert repeated prompts into templates and scheduled automations so common coding loops become faster to restart.
Use transparent tool events and run history to see what happened before you accept, revise, or discard the agent output.