Management layers on top of coding work. Agents (Claude Code, Codex, Copilot, Gemini) write the code; the layers above assign, gate, review and account for it.
- Batch remote and local coding-agent work with task management in one place.
- Reduces cognitive load: task state, progress and evidence are visible instead of remembered.
- A management layer over autonomous agent runs — assign, gate, review, account.
- Works with your existing CLI subscriptions, or bring your own API keys.
- Runs fully local or distributed — your choice per project.
| Repository | Package | What it is |
|---|---|---|
| agent-studio | — | The workbench: task board, agent pipelines, project wikis, quota & release management. .NET backend, Angular frontend. |
| quality-studio | — | Agent-driven, layered code review with the quality verdict persisted next to the code. |
| runner | CodingAgentRunner (NuGet) |
Run coding-agent CLIs from .NET: hardened spawning, streamed events, lifecycle & quota tracking. |
| token-economy | TokenEconomy (NuGet, first release upcoming) |
Token economics for LLM coding agents: dated model pricing, cost computation, token-efficiency model selection. |
| chat | coding-agent-chat (npm) |
Angular UI that renders an agent's event stream as a readable conversation. |
Apache-2.0 · agent-orchestrator.dev