Cloud CLI

· #454 most-used

Run AI coding agents in persistent cloud environments, triggered by any workflow

ProductivityDeveloperAutomationCloud & InfrastructureMonitoring & Alerts

Cloud CLI is a managed cloud environment platform built for AI coding agents. It runs Claude Code, Codex, Cursor CLI, and Gemini CLI in persistent isolated containers that stay running even when your laptop closes. Connect it to Actionist and your agents can provision development environments, execute AI coding tasks from any workflow trigger, stop idle containers automatically, retrieve execution logs, and orchestrate full software development lifecycles — from Jira ticket to merged pull request — without a developer touching a keyboard.

Average time saved
10 hours
per person · per month
≈ 1 workdays back

Eliminates manual work. Agents eliminate manual environment provisioning, idle environment monitoring, nightly maintenance runs, and bug reproduction setup that development and operations teams repeat every week.

Schedule

What your Cloud CLI agent runs on autopilot

A week of scheduled jobs your Actionist agent will execute on your behalf.

28Scheduled jobs
7Agents at work
24/7Always on
Agents
TueThu
Tue
Wed
Thu
7a
8a
9a
10a
11a
12p
1p
2p
3p
4p
5p
6p
Multi-app workflows

Cloud CLI × every other app you use

End-to-end automations that span multiple apps — each one a real business outcome.

6Workflows
6Apps spanned
~45 hrsSaved / week
6Personas served
For engineering
Featured2 apps

GitHub issue auto-implemented by AI agent

When a GitHub issue is labelled 'ai-ready', the agent confirms the environment is live, runs Claude Code against the issue description, retrieves the output, and posts the IDE link and pull request back to the GitHub issue — implementation work is in progress within about a minute, without a developer opening their laptop.

~23 hrs

Time saved for your team — every week, on autopilot

The flow
Trigger·When a GitHub issue receives the ai-ready label
Result
Execute Agent with ticket title and description as promptPost IDE deep link and PR link as comment on the issue
The win
Saved per run
~2 hrs
Runs / week
~15×
Engineers review agent-written code instead of writing boilerplate
Driven byOperations Agent
ROI

Savings

What your team gets back — two angles: what you stop doing manually, and what that's worth.

Without Actionist

What you do manually today

With Actionist

What your agent runs for you

  • Sales
    120 min / week
    Manual demo environment setup

    Solutions engineers spend 30–60 minutes manually provisioning, configuring, and seeding demo environments for each enterprise prospect — often done ad-hoc the morning of the call.

    Sales Agent
    0 min
    Agent builds and posts POC environment before the call

    When a new enterprise prospect is added to CRM, the agent creates a Cloud CLI environment, seeds it with a demo project via Execute Agent, and shares the live environment link with the account executive — a working prototype is ready before the discovery call.

  • Marketing
    90 min / week
    Manual developer documentation maintenance

    The developer relations team manually runs code samples against the current API after each release to check accuracy — a process that takes hours and often gets skipped under release pressure.

    Marketing Agent
    0 min
    Agent regenerates and verifies code samples automatically

    Every week the marketing agent calls Execute Agent to run all public code samples in the docs environment against the current API, posting any failures to the dev docs channel before they reach developers in the wild.

  • Customer Support
    60 min / week
    Manual bug reproduction setup

    Support engineers manually set up local environments to reproduce customer-reported bugs — pulling the right version, installing dependencies, running the steps — before they can even confirm the bug is real.

    Customer Support Agent
    0 min
    Agent reproduces bugs automatically before triage

    When a bug report arrives, the agent creates an isolated environment, runs the reproduction steps, and posts a confirmed reproduction result to the GitHub issue — engineering triage starts with facts rather than a vague description.

  • Human Resources
    60 min / week
    Manual developer onboarding environment setup

    IT or senior engineers spend time manually provisioning development environments for new hires — running setup scripts, troubleshooting dependency issues, and confirming the environment works, often on the new hire's first day itself.

    Human Resources Agent
    0 min
    Agent provisions verified environment before day one

    When a new engineer is added to the onboarding sheet, the agent clones the golden-path environment, runs the setup verification, and sends the new hire their workspace link — all before they arrive on their first day.

  • Finance
    30 min / week
    Unmanaged idle cloud environment spend

    Development environments left running over weekends and evenings accumulate unnecessary compute costs — no one tracks idle time systematically, and the cost only surfaces when the monthly bill arrives.

    Finance Agent
    0 min
    Agent stops idle environments nightly to prevent waste

    Every evening the finance agent checks for environments idle more than two hours, stops them automatically, and logs the cost impact — cloud spend tracks actual usage rather than developer oversight.

  • Operations
    90 min / week
    Manual scheduled maintenance tasks

    Platform engineers manually run dependency update checks, security scans, and health checks across development environments on an ad-hoc schedule — tasks that slip under sprint pressure and surface as incidents.

    Operations Agent
    0 min
    Agent runs nightly maintenance across all environments

    The operations agent runs dependency updates, security scans, and test suite checks every night via Execute Agent, delivering a consolidated maintenance report to Slack by morning standup — planned maintenance becomes invisible background work.

  • Legal
    20 min / week
    Manual offboarding access revocation

    IT and security teams manually identify and delete developer environments during offboarding — a process that relies on runbooks, cross-team coordination, and is often completed days after the employee's last day.

    Legal Agent
    0 min
    Agent archives logs and deletes access within minutes of offboarding

    When an offboarding record is completed, the agent archives environment logs and deletes all workspaces within about a minute, writing a dated audit record to the compliance log — access revocation is complete before the departing employee's laptop is collected.

+ 100s of other Cloud CLI automations
Average time saved
47 hrs / person / month
Calculator

Calculate what your team saves

Team size
5 people
Hourly rate
$75 / hr
Hours saved / week
13
Hours saved / year
625
Annual ROI
$46,875

Based on Cloud CLI's typical team usage — the visible tasks plus a few other automations the agent runs: ~2.5 hrs / person / week of admin work automated.

Connect

How to plug Cloud CLI into Actionist

Pick the connection method that suits your environment.

Authenticate Actionist with a Cloud CLI API key generated in your account settings. The key gives the agent access to all environment and agent execution APIs.

1
Open the API keys page

Sign in to cloudcli.ai and go to cloudcli.ai/api-keys. Click Generate key.

2
Copy the key before closing the dialog

Copy the key immediately — Cloud CLI shows it only once at creation. Store it in a password manager.

3
Paste into Actionist and test

Paste the key into the API key field in Actionist and click Test connection. The agent confirms access by listing your environments.

Credentials you'll need
API key*
cloudcli.ai/api-keys — generate a key and copy it immediately, it is shown only once
Actions

12 actions your agent can call

Read and write operations available to your Actionist agent.

Triggers

0 events your agent can react to

Events your agent watches for, and the actions it kicks off in response.

This app has no triggers yet.
MCP servers

MCP servers that work with Cloud CLI

Connect Actionist to MCP servers built for or around this app.

Redis MCP Cloud

Manage Redis Cloud resources using natural language. Create databases, monitor subscriptions, and configure cloud deployments.

dbt MCP
Official

Official MCP server for dbt providing integration with dbt Core/Cloud CLI, project metadata discovery, model information, and semantic layer querying.

IBM Cloud CLI MCP

IBM Cloud CLI MCP server for managing IBM Cloud resources and infrastructure.

Bitbucket Cloud CLI

Bitbucket Cloud CLI and MCP server. Manages workspaces, repos, PRs, pipelines, issues, and source code.

FAQs

Questions about Cloud CLI + Actionist

How does Actionist connect to Cloud CLI?
Go to the Apps tab in Actionist, find Cloud CLI, and click Connect. You will be prompted to enter your Cloud CLI API key, which you generate at cloudcli.ai/api-keys. Cloud CLI shows the key only once at creation time — copy it immediately and store it securely. Once pasted into Actionist, the agent runs a test call to confirm the connection before any automation runs.
Can Actionist instruct an AI coding agent to implement a feature automatically?
Yes. The Execute Agent action accepts a plain-text prompt describing the task — for example, 'Implement the user authentication module from the attached spec and open a pull request.' The agent picks up the environment you specify, runs Claude Code (or Codex, Cursor CLI, or Gemini CLI) inside the isolated container, and returns the output, IDE deep links, and SSH commands so a developer can pick up exactly where the agent left off.
How are Cloud CLI environments priced and how does Actionist manage them?
Cloud CLI environments run in isolated containers on CloudCLI's managed infrastructure. Each environment gets its own filesystem, SSH access, and a unique subdomain. Environments are priced per slot (Hobby: 1 environment at €7/month; Growth: 5 environments at €20/month, expandable at €5 each). Actionist can start, stop, and delete environments via the API, so you only pay for active slots — idle environments can be stopped programmatically between agent tasks.
Which AI coding agents does Cloud CLI support?
Cloud CLI supports Claude Code, Cursor CLI, OpenAI Codex, and Gemini CLI inside its containerized environments. You specify which agent to use when calling Execute Agent. Actionist passes the prompt and environment ID — Cloud CLI routes the task to the correct agent runtime inside the container. All agents run with persistent file state, meaning a task left incomplete by one agent run can be resumed in the same environment.
Do Cloud CLI environments retain state between Actionist runs?
Environments persist between sessions — stopping an environment suspends compute billing while retaining the filesystem. The workspace directory (/workspace/) and all project files remain intact. When Actionist starts the environment again for the next task, the agent picks up exactly where it left off, including Git history, installed dependencies, and any in-progress work. This is the key architectural difference from ephemeral CI runners.
What are the most common Actionist automations using Cloud CLI?
Common patterns: (1) ticket-to-code — when a new GitHub issue or Jira ticket is created, Actionist executes an agent task in Cloud CLI to implement the feature or reproduce the bug; (2) scheduled maintenance — nightly agent tasks that update dependencies, run security scans, or generate reports; (3) on-demand code review — when a pull request opens, an agent reviews the diff and posts comments; (4) documentation generation — when a module changes, an agent rewrites the corresponding docs section. All of these require only the API key connection and the relevant trigger app.
Can a developer continue work after an Actionist agent finishes a Cloud CLI task?
Yes. Cloud CLI environments support SSH access and generate IDE deep links (VS Code and Cursor) automatically when an agent task completes. Actionist can post these links to Slack or a Jira comment so a developer can open the environment in their preferred IDE or SSH in and run `claude -r` to resume the session. The handoff is seamless — the agent's partial work is right there, ready to inspect or continue.
Can Cloud CLI be self-hosted for data residency requirements?
Cloud CLI is open source (GPL-3 licensed, built on ClaudeCodeUI with over 11,000 GitHub stars) and supports self-hosting on your own infrastructure. If your organisation has data residency requirements, you can run the Cloud CLI server internally and point Actionist at your self-hosted API endpoint. The API contract is the same, so all Actionist actions work identically against a self-hosted instance.