Development
Plugin
agent-orchestrator
The orchestration layer for parallel AI coding agents An Agentic IDE that supervises parallel AI coding agents in isolated workspaces, with complete control and automatic feedback loops from CI failures, review comments, and merge conflicts.
Plugin Signal85/100Strong
maintained100
capable58
documented100
adopted84
Install
$ npx -y skills add Untrivial-ai/agent-orchestrator --agent claude-code
Run the curl in your terminal, the rest in Claude Code.
What's inside
Skills
2/ao-desktop-devLaunch, restart, or troubleshoot the real AO Electron desktop app from this repository; run a checkout against isolated or real local AO data; combine PR branches for local UI review; and diagnose stale Electron processes, port conflicts, or preload bridge mismatches. Use/bug-triageTriage bugs reported in chat/issues, search for duplicates, file or update GitHub issues with full context, and push fix PRs.
Commands
13/browserInspect and control the current AO session's target-isolated browser. The desktop app must be open. The agent and user share the same live page, cookies, navigation state, and `WebContentsView`; the runtime remains usable while the Browser panel is hidden. Tabs in this worker/doctorRun local AO health checks. Use this to diagnose setup problems or verify the environment is correctly configured./importImport reads the legacy Agent Orchestrator flat-file store (`~/.agent-orchestrator`) read-only and ports its projects and per-project settings into the rewrite database. Legacy files are never modified, and a re-run skips rows that already exist, so it is safe to run more than/orchestratorManage orchestrator sessions./previewOpen a URL or workspace file in the desktop browser panel for the current session, or start a deterministic session-owned dev server from an existing `.ao/launch.json`./projectManage projects: register repos, inspect, configure per-project settings, and remove.
Show all 13 commandsShow fewer
/reviewManage AO code reviews of a worker's PR./sendSend a message to a running agent session. Use this to correct or direct a live agent mid-stream without killing and respawning it./sessionManage agent sessions: list, inspect, rename, kill, restore, clean up, and claim PRs./spawnSpawn a worker agent session in a registered project. The session runs the chosen agent in a fresh git worktree. Register the project first with `ao project add`./startFetch (if needed) and open the Agent Orchestrator desktop app. The desktop app owns the daemon, state, and updates. `ao start` no longer runs a daemon: it resolves the installed app (or downloads the latest release), opens it, and exits./statusShow AO daemon status. Use this to verify the daemon is up and check which port it is bound to./stopStop the AO daemon.
README
1Agent Orchestrator
The orchestration layer for parallel AI coding agents
English · 简体中文 · 日本語 · 한국어 · Español · Français · Deutsch · Português (Brasil)
An Agentic IDE that supervises parallel AI coding agents in isolated workspaces, with complete control and automatic feedback loops from CI failures, review comments, and merge conflicts.
What is Agent Orchestrator?
Agent Orchestrator is a meta-harness agent IDE for running AI coding agents in parallel. It gives agents like Claude Code, Codex, Cursor, Kimi Code, opencode, and others a shared workspace where their sessions, chat or terminal interfaces, branches, pull requests, and feedback loops can be supervised from one place.
The agents still do the coding. AO provides the harness around them: isolated workspaces, live terminal access, session state, PR awareness, and automatic loops that send CI failures, review comments, and merge conflicts back to the right agent. Instead of manually coordinating a pile of agent terminals, AO turns parallel agent work into a managed workflow.
Why Agent Orchestrator?
AI coding agents become much more useful when they can work in parallel, but parallel work gets messy quickly. Branches overlap, terminals get lost, CI failures need follow-up, review comments need replies, and merge conflicts have to reach the right worker.
Agent Orchestrator is built to keep that loop visible and manageable. It helps you:
- Start multiple agents from the same project without mixing their work
- Keep every session in a separate git worktree
- See which agents are working, waiting, finished, or blocked
- Route CI failures, review comments, and merge conflicts back to the right session
- Use different agent CLIs through one common supervisor
How it works
At a high level, Agent Orchestrator follows a simple loop:
- Add a project you want agents to work on.
- Start one or more sessions from the desktop app or CLI.
- AO creates an isolated git worktree for each session.
- AO launches either the agent's terminal UI or a structured Chat controller, according to the session's selected interface.
- The local daemon watches session state, controller activity, pull requests, CI, and review feedback.
- The desktop app and CLI show the current state and let you send follow-up instructions to the right session.
The result is a local control layer for agentic coding: agents still do the coding, while Agent Orchestrator keeps their workspaces, status, terminals, and feedback loops organized.
Features
The desktop app is the main control surface: projects on the left, active sessions in the center, and the selected session's Chat or terminal interface, pull request state, review runs, and browser preview in the inspector.
Supported Agents
AO ships adapters for 25 worker agent harnesses:
Reviewer agents are configured separately. The current reviewer harnesses are:
Reviewer panes stay interactive. AO opens a fresh reviewer process for each pass so the selected harness's current task context, permissions, and environment are applied at launch. Pi reviewers run with project/user resource discovery and built-in tools disabled; AO loads an extension from its own data directory that exposes only read-only checkout inspection plus structured GitHub review posting and ao review submit. Agy, Continue, Devin, Droid, Goose, Kimchi, Kimi, Qwen, and Vibe are experimental host-trusted reviewers: their native modes and settings do not provide OS isolation, and some reviewers receive autonomous settings so they can complete review tasks. Grok, Crush, Auggie, Cline, and Autohand are experimental user-approved reviewers: AO supplies a hardened reviewer role and leaves native permission prompts enabled rather than passing broad unattended flags.
If it runs in a terminal, it runs on Agent Orchestrator.
Use the agent interface that fits the moment: structured Chat or its native terminal UI.
Install
Download the latest desktop build for your platform:
| Platform | Download |
|---|---|
| macOS (Apple silicon) | Download |
| macOS (Intel) | Download |
| Windows | Download |
| Linux (AppImage) | Download |
| Linux (Debian/Ubuntu) | Download |
| Linux (Fedora/RHEL) | Download |
After installing, open Agent Orchestrator and point it at the repository you want AO to manage. The desktop app runs the daemon for you, so no CLI is required. Installed desktop builds check for updates on launch and periodically while the app is running. See the installation guide for agent CLI setup and troubleshooting.
npm still works but is no longer recommended. 0.10.0 is the final version published to npm, and the @aoagents/ao package is frozen and will not receive further updates. It stays available for existing users who have the ao CLI on their PATH; ao start fetches and opens the same desktop build linked above. For any new setup, prefer the desktop download.
npm install -g @aoagents/ao
ao start
Witness AO's Journey on X
Documentation
| Document | Start here when you need |
|---|---|
| docs/architecture.md | Backend mental model, lifecycle, persistence, CDC, status derivation, and daemon boundaries. |
| docs/backend-code-structure.md | Package ownership and where each backend concern belongs. |
| docs/cli/README.md | CLI behavior and daemon route mapping. |
| docs/development.md | Prerequisites, build steps, running tests, and troubleshooting for local development. |
| docs/STATUS.md | What currently ships on main and what remains in flight. |
| docs/stack.md | Library, runtime, and dependency decisions. |
Telemetry
Agent Orchestrator's Electron renderer sends anonymous usage events to PostHog for reliability and product understanding. PostHog session recording is disabled by default; if a time-boxed investigation enables it, local paths and local URLs are redacted before transmission. Set VITE_AO_POSTHOG_KEY to an empty string before building to disable transmission. See docs/telemetry.md.
License
Apache License 2.0. See LICENSE.
File tree
1626Untrivial-ai/agent-orchestratorView on GitHub
.agents/
skills/
ao-desktop-dev/
agents/
openai.yaml
SKILL.md
bug-triage/
scripts/
push_fix_to_github.py
SKILL.md
.claude/
skills/
ao-desktop-dev/
SKILL.md
bug-triage/
scripts/
push_fix_to_github.py
SKILL.md
.envrc
.gitattributes
.github/
.gitleaks.toml
actions/
macos-signing-setup/
action.yml
ISSUE_TEMPLATE/
bug_report.yml
config.yml
feature_request.yml
pull_request_template.md
workflows/
build-artifacts.yml
cli-e2e.yml
deploy-landing.yml
desktop-testing.yml
e2e-gate-tests.yml
feature-release-cleanup.yml
feature-release.yml
frontend-release.yml
frontend.yml
gitleaks.yml
go.yml
mac-update-e2e.yml
mobile.yml
react-doctor.yml
release-latest-guard.yml
testing-build.yml
.gitignore
AGENTS.md
assets/
ao-dashboard-preview.png
ao-logo.svg
first.png
image.png
second.png
third.png
tweet1.png
tweet2.png
tweet3.png
backend/
.golangci.yml
cmd/
ao/
main.go
gencodexproto/
main.go
genspec/
main.go
e2e/
chat_compaction_test.go
chat_composer_test.go
chat_conversation_test.go
chat_history_test.go
chat_lifecycle_test.go
chat_mode_test.go
chat_output_diff_test.go
chat_provider_state_test.go
chat_steer_test.go
chat_usage_test.go
harness_test.go
go.mod
go.sum
internal/
adapters/
agent/
activitydispatch/
dispatch_test.go
dispatch.go
activitystate/
activitystate_test.go
activitystate.go
agentbase/
agentbase_test.go
agentbase.go
agy/
activity_test.go
activity.go
agy_test.go
agy.go
auth.go
hooks.go
install.go
aider/
aider_test.go
aider.go
auth_test.go
auth.go
install.go
amp/
amp_test.go
amp.go
auth_test.go
auth.go
hooks.go
install.go
auggie/
auggie_test.go
auggie.go
auth_test.go
auth.go
install.go
authprobe/
authprobe_test.go
authprobe.go
autohand/
auth_test.go
auth.go
autohand_test.go
autohand.go
hooks.go
install.go
binaryutil/
binaryutil_test.go
binaryutil.go
claudecode/
activity_test.go
activity.go
claudecode_test.go
claudecode.go
continuation.go
hooks.go
install.go
cline/
auth_test.go
auth.go
cline_test.go
cline.go
hooks.go
install.go
codex/
activity_test.go
activity.go
codex_test.go
codex.go
continuation.go
hooks.go
install.go
terminal_activity_test.go
terminal_activity.go
continueagent/
auth_test.go
auth.go
continueagent_test.go
continueagent.go
install.go
copilot/
auth.go
copilot_test.go
copilot.go
hooks.go
install.go
crush/
activity_test.go
activity.go
auth_test.go
auth.go
crush_test.go
crush.go
hooks.go
install.go
cursor/
auth_test.go
auth.go
cursor_test.go
cursor.go
hooks.go
install.go
devin/
auth_test.go
auth.go
devin_test.go
devin.go
hooks.go
install.go
droid/
activity_test.go
activity.go
auth_test.go
auth.go
droid_test.go
droid.go
hooks.go
install.go
fake/
fake_test.go
fake.go
goose/
auth.go
goose_test.go
goose.go
hooks.go
install.go
grok/
auth_test.go
auth.go
grok_test.go
grok.go
install.go
hooksjson/
hooksjson_test.go
hooksjson.go
hookutil/
hookutil_test.go
hookutil.go
kilocode/
assets/
ao-activity.ts
auth_test.go
auth.go
hooks.go
install.go
kilocode_test.go
kilocode.go
kimchi/
activity_test.go
activity.go
auth_test.go
auth.go
hooks.go
install.go
kimchi_fifo_test.go
kimchi_test.go
kimchi.go
kimi/
auth_test.go
auth.go
hooks.go
install.go
kimi_test.go
kimi.go
kiro/
auth.go
hooks.go
install.go
kiro_test.go
kiro.go
modelcatalog/
catalog_test.go
catalog.go
muse/
activity_test.go
activity.go
auth_test.go
auth.go
hooks.go
install.go
muse_test.go
muse.go
testdata/
active_generation.txt
awaiting_user_input_compact.txt
awaiting_user_input.txt
idle_composer.txt
nativeconfig/
nativeconfig.go
opencode/
activity_test.go
activity.go
assets/
ao-activity.ts
hooks.go
install.go
opencode_test.go
opencode.go
pi/
auth_test.go
auth.go
install.go
pi_test.go
pi.go
primeagent/
activity_test.go
activity.go
assets/
ao-activity.ts
cancellation_unix_test.go
hooks_test.go
hooks.go
primeagent_test.go
primeagent.go
qwen/
auth_test.go
auth.go
hooks.go
install.go
qwen_test.go
qwen.go
registry/
registry_test.go
registry.go
terminalui/
composer_test.go
composer.go
vibe/
activity_test.go
activity.go
auth.go
hooks.go
install.go
vibe_test.go
vibe.go
chatdriver/
acp/
client.go
config_options.go
conversation.go
driver_test.go
driver.go
history.go
process_unix.go
process_windows.go
process.go
session_setup.go
skills.go
steer.go
claudeacp/
driver_test.go
driver.go
live_test.go
codexappserver/
codexproto/
gen.go
protocol_test.go
protocol.gen.go
conformance_test.go
conversation.go
diff_test.go
diff.go
driver_test.go
driver.go
history_test.go
history.go
live_test.go
normalize_ingest_test.go
normalize_test.go
normalize.go
rpc_test.go
rpc.go
skills_test.go
skills.go
steer_test.go
steer.go
droidacp/
driver_test.go
driver.go
live_test.go
nativeacp/
driver_test.go
driver.go
opencodeacp/
driver_test.go
driver.go
live_test.go
processenv/
processenv_test.go
processenv.go
registry/
registry_test.go
registry.go
container/
dockerreap/
reap_test.go
reap.go
registry.go
reviewer/
agentrestore/
agentrestore.go
agy/
agy_test.go
agy.go
aider/
aider_test.go
aider.go
amp/
amp_test.go
amp.go
auggie/
auggie_test.go
auggie.go
autohand/
... 1600 moreShowing a partial view of a very large repo.
FAQ
agent-orchestrator is a Claude Code plugin with 2 hand-picked skills for development work, indexed on Flowy. Install it with the command on its page. It includes ao-desktop-dev, bug-triage. Its skills do not fire on their own yet. Request auto-invocation to have Flowy route them as you prompt. Free and open source.
- What is agent-orchestrator?
- The orchestration layer for parallel AI coding agents An Agentic IDE that supervises parallel AI coding agents in isolated workspaces, with complete control and automatic feedback loops from CI failures, review comments, and merge conflicts.
- How do I install agent-orchestrator?
- Run these in Claude Code: npx -y skills add Untrivial-ai/agent-orchestrator --agent claude-code. Then prompt normally.
- Does agent-orchestrator auto-invoke its skills?
- Not yet. It is indexed on Flowy as a plain plugin. Request auto-invocation on its page and Flowy will route its skills for you as you prompt.
- Is agent-orchestrator free?
- Yes. Flowy is free and open source, with nothing gated. You can read every skill in full before you install.