CLI Agent Orchestrator — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow AWS integration and orchestration changes as cloud tooling evolves

Reviewed Oct 5, 2026 · Documentation review; complete setup workflow not independently tested

Best fit

Teams coordinating multiple authenticated coding-agent CLIs in parallel or sequence.

What we know

A local orchestrator for multiple coding-agent CLIs with supervisor coordination, web UI, CLI, MCP and plugin/control-plane surfaces.

Maintenance / activity evidence

Observed Oct 5, 2026 (UTC)fix(copilot cli): detect idle on Copilot CLI via the rendered pane ( …

Tracking evidence observed Oct 5, 2026. These dated signals do not establish current maintenance. Reviewed enrichment confidence: high.

Activity source · commits ↗Activity source · releases ↗

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Full provider-CLI capabilities subject to CAO restrictions. Writes: Coding-agent workspace writes; Native provider tools.
Provider CLI auth
Sensitive access: Repositories, Provider credentials/config

CAO orchestrates powerful coding agents; tool restrictions, project scope and provider credentials are the primary boundaries.

Security source · repository notice ↗Security source · policy file ↗

Full authority, access and evidence limits are available in Review context and Access & risk.

Popularity

1,393 repository stars · 1.1k package downloads / 30 days

Stars observed Oct 6, 2026. Downloads observed Oct 6, 2026.

All limitations / Review context

Which AWS and MCP orchestration paths are currently documented?

The bounded assessment reviewed the official README, repository notices, security policy path, and security page for known material public-tracking blockers. No such blocker was identified. This is a public-tracking assessment, not an installation safety certification.

CAO orchestrates powerful coding agents; tool restrictions, project scope and provider credentials are the primary boundaries.
Full provider-CLI capabilities subject to CAO restrictions.
Provider CLI auth
Workspace files, terminal output and optional memory may be processed or persisted.
Reviewed write actions: Coding-agent workspace writes; Native provider tools
Sensitive access: Repositories, Provider credentials/config
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
No independent runtime, installation, or comprehensive security audit was performed.
No installation, runtime, client matrix, or complete tool-contract audit was performed in this review.
Requires provider CLI authentication.
Agents retain broad native capabilities inside sessions.
Security depends on workspace and tool restrictions.

Original activity observations

2026-10-05T13:57:53Z: fix(copilot cli): detect idle on Copilot CLI via the rendered pane ( …

Technical identity

github.com/awslabs/cli-agent-orchestrator

Exact implementation source: https://github.com/awslabs/cli-agent-orchestrator ↗
Reviewed resource type: Host
Repository identity: awslabs/cli-agent-orchestrator
Package identity: pypi:cli-agent-orchestrator
Service endpoint: http://localhost:9889

Evidence & Sources

Evidence basis: Recorded source evidence
Setup runtime verification: Not independently runtime-tested by MCP100 for this setup workflow.

Reviewed Oct 5, 2026 · Latest source observation Oct 6, 2026

GitHub repository ↗
awslabs/cli-agent-orchestrator
Recorded repository material and project history.
Evidence source ↗
pypi.org/project/cli-agent-orchestrator/
Source material retained for this review.
Includes references whose attribution is not established.

Setup · Watchlist

Host / Desktop app setup

Publisher instructions.

Source evidence observed Oct 13, 2026.

Additional setup details required — see the specific gaps below.

Download / install
Publisher instructions
CLI Agent Orchestrator is a multi-agent supervisor, not a standalone MCP server. Install via official uv tool install, check cao --help, choose a supported provider and explicitly start a local supervisor.
uv tool install git+https://github.com/awslabs/cli-agent-orchestrator.git@main --upgrade
Advanced configuration · self hosted install or run
self hosted install or run · Review before use
cao install code_supervisor cao-server
Configure application / integrations
Publisher instructions
CAO is a supervisor for coding agents, not an MCP server to register. Requires uv and supported agent CLIs; agent plugins are gated off by default. The supervisor runs locally after a selected provider CLI is available; follow the first-launch recipe in the official README. Its optional operations MCP is different from installing a CAO CLI.
Check connection
MCP100 setup guide
Use cao --help to verify CLI installed, then create the supported supervisor profile, start cao-server and inspect the provider-specific startup path. Do not automatically run unreviewed agent commands or start production jobs.
cao --help
Disconnect / uninstall / revoke
Publisher instructions
This is an agent supervisor, not a standalone MCP server. Use publisher `cao shutdown --session <session-name>` to stop a specific agent session. Removing its installed executable is separate; do not run `cao shutdown --all` merely to disconnect one session.
cao shutdown --session <session-name>

Official Documentation & Sources

What it does

A local orchestrator for multiple coding-agent CLIs with supervisor coordination, web UI, CLI, MCP and plugin/control-plane surfaces.

Capability: AI & Reasoning
Resource type: Host
Industry: Software & Technology
Best fit: Teams coordinating multiple authenticated coding-agent CLIs in parallel or sequence.

Representative uses

Run specialist coding agents
Launch supervisors over workers
Coordinate through UI, CLI or MCP
Deploy agent fleets in documented Kubernetes examples

Limitations

Requires provider CLI authentication.
Agents retain broad native capabilities inside sessions.
Security depends on workspace and tool restrictions.

Reviewed access

Access & risk

Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.

Authentication: Provider CLI auth
Permissions: Full provider-CLI capabilities subject to CAO restrictions.
Data handling: Workspace files, terminal output and optional memory may be processed or persisted.
Cost: usage_based
Risk and cautions: CAO orchestrates powerful coding agents; tool restrictions, project scope and provider credentials are the primary boundaries.

Access scope and cautions

Write actions: Coding-agent workspace writes Native provider tools
Sensitive access: Repositories Provider credentials/config
Risk level: high

FAQ

Based on reviewed evidence as of 2026-10-06. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Run specialist coding agents; Launch supervisors over workers; Coordinate through UI, CLI or MCP; Deploy agent fleets in documented Kubernetes examples.

What setup and authentication are documented?

A complete connection procedure is not verified. Recorded endpoint: http://localhost:9889. Recorded transport: other. These facts do not supply missing installation commands, credentials or a verified authentication procedure. Authentication: Provider CLIs retain native authentication.

Can it modify data or perform write operations?

Documented write actions include: Coding-agent workspace writes; Native provider tools. Recorded permission boundary: Full provider-CLI capabilities subject to CAO restrictions.

What data-handling boundary should I consider?

Workspace files, terminal output and optional memory may be processed or persisted. Recorded sensitive-access context: Repositories; Provider credentials/config.

What limitations and risks should I check before use?

CAO orchestrates powerful coding agents; tool restrictions, project scope and provider credentials are the primary boundaries. Requires provider CLI authentication. Agents retain broad native capabilities inside sessions. Security depends on workspace and tool restrictions. No independent runtime, installation, or comprehensive security audit was performed.

What cost model is recorded?

The reviewed cost model is usage based. This classification does not establish current prices, quotas or the cost of every downstream operation. Check the applicable service terms before authorizing paid actions.

Is this an official MCP?

AWS Labs is listed as First-party MCP in MCP100’s current evidence.