SkillFoxx

Type a name, a task or a service

Match with Foxx AI

AEnvironment Deploy skill

A Claude Code skill that rolls out AEnvironment platform environments as temporary instances or permanent services

  • 322 ★ on the repository
  • Last commit
  • Apache-2.0
SkillOfficial

Broad access

Why this level

  • The scripts create and delete instances, services and storage in your cluster, and deleting with `delete_storage=True` wipes data
  • The registry password goes in the `--registry-password` flag and service keys in `--env-vars`, so they land in command history
  • The local-build workflow runs Docker and pushes an image to a registry
All reasons and checks

inclusionai/aenvironmentAuthor:inclusionAI

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/aenvironment-deploy

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • The scripts create and delete instances, services and storage in your cluster, and deleting with `delete_storage=True` wipes data
  • The registry password goes in the `--registry-password` flag and service keys in `--env-vars`, so they land in command history
  • The local-build workflow runs Docker and pushes an image to a registry

This is third-party code. Review the repository files before installing.

What it does

AEnvironment is an environment platform for training and running agents that Ant Group uses internally. Its environments are described through MCP, and TAU2, a mini terminal and TerminalBench come built in. The `aenvironment-deploy` skill takes over rollout: Claude Code configures the `aenv` CLI, registers the environment, creates an instance or a service and injects environment variables. The skill walks through three workflows. In the first, a Docker image is built locally, registered in EnvHub and deployed. In the second, a prebuilt image is registered. In the third, an already registered environment is deployed, which is the shortest path. Each has its own Python script: `deploy_with_local_build.py`, `deploy_with_existing_image.py` and `deploy_existing_env.py`. An instance lives temporarily and opens by IP and port, with its lifetime set by `--ttl`, 24 hours by default. A service runs permanently, opens by domain name, allows several replicas and gets persistent storage with `--enable-storage`, but then it stays at one replica. The skill lists, updates and deletes resources through `scripts/aenv_operations.py`.

Who it is for. Researchers and engineers who train and test agents on their own AEnvironment deployment and want to roll out environments from Claude Code.

Good fit when

  • The `game-2048@1.0.6` environment is already registered, and you need a one-hour instance to test an agent
  • An application needs a permanent service with a domain and 10 GB of storage
  • A Docker image sits in your registry and has to be registered in EnvHub and deployed at once
  • Stale instances and services need to be listed and deleted, keeping or wiping the storage

Not a fit when

  • You have no deployed API service or EnvHub: the scripts take their addresses and create nothing without them
  • You only need a local check of an environment: `aenv run` is enough and starts an MCP server in the project folder
  • The agent is not Claude Code: installation is described with the `claude skill install` command

Example request

Deploy game-2048@1.0.6 as an instance with a one-hour lifetime

Limitations

  • The skill manages deployment but does not install the platform itself. You must know the API service and EnvHub addresses in advance.
  • The local-build workflow needs Docker and registry credentials: host, user, password and namespace.
  • The CLI installs with `pip install aenvironment --upgrade`, and the skill needs Python.
  • The sample service address in SKILL.md points at an internal Ant Group cluster, so your address depends on your own installation.

How to disable. Delete the `aenvironment-deploy` skill folder from Claude Code's skills directory. Deployed instances and services keep running, and you remove them separately through `aenv` or `scripts/aenv_operations.py`.

FAQ

How do I install AEnvironment Deploy in Claude Code?

Download `aenvironment-deploy.skill` from the latest release with `curl -L` and run `claude skill install aenvironment-deploy.skill`. The skill's source sits in the repository under `.claude/aenvironment-deploy`.

How does an instance differ from a service?

An instance is temporary with a lifetime set by `--ttl` (24 hours by default), IP and port access, one replica and no storage. A service runs permanently, opens by domain, allows several replicas and gets storage with `--enable-storage` at one replica.

Which workflow fits an environment that is already registered?

The third: the `deploy_existing_env.py` script takes the environment name and version as `myagent@1.0.0`, the API service and EnvHub addresses, the deployment type and the owner name. Image building and the registry take no part.

How do I pass keys into the environment?

With the `--env-vars` flag and a JSON dictionary, for example holding model and news API keys. The skill injects them into the instance or service on creation.

How do I delete a service together with its storage?

Through `scripts/aenv_operations.py`: the `delete_service` method with `delete_storage=True` removes both the service and its storage. Without that argument the storage stays.

Discussion

No questions or reviews yet.

Skills

Docker Skills

docker-skills

Official

Official Docker skills for agents: Dockerfile and builds, Compose, Sandboxes, Docker Agent and a skill that confirms irreversible actions

SkillMinimal access615
Editors’ pick

Isolated Linux microVMs for agent code: the official Vercel skill on the SDK, network rules and snapshots.

SkillExtended access219

AgentBox runs Claude Code, Codex and Open Code in separate virtual machines on your PC, a self-hosted server or a cloud.

CLIExtended access570

Cloud.ru Skills for AI Agents

Cloud.ru Skills для AI-агентов

Official

Official Cloud.ru skills: account setup, Foundation Models, ML Inference, virtual machines, RAG and Evolution AI Agents

SkillBroad access7

31 DevOps skills for Claude Code and Codex: config generators, validators and a Kubernetes debugger for Terraform, Helm, CI and Docker

SkillExtended access332

A Docker sandbox for coding agents: a Debian image with Copilot CLI, Codex and Pi, and SSH, RDP and Docker enabled by flags.

CLIExtended access154

README in short

The page lays out the "everything is an environment" idea, built-in environments, mini-program and TAU2 reinforcement-learning examples and the agent-as-environment mode. The deploy skill has its own section with installation and three workflows, followed by SDK setup and how to define environment tools and functions.

README badge

Original SKILL.md

---
name: aenvironment-deploy
description: Deploy sandboxed environment instances and services using AEnvironment. Use when deploying agent instances, web services, or applications to AEnvironment sandbox infrastructure. Supports three workflows - (1) Build image locally and deploy, (2) Register existing image and deploy, (3) Deploy from registered environment. Handles instance deployment (temporary, IP-based access for agents) and service deployment (persistent, domain-based access with storage for apps).
---

# AEnvironment Deploy

Automate deployment of sandboxed environment instances and services using the AEnvironment platform.

## Overview

AEnvironment provides isolated sandbox environments for running agents and applications. This skill handles the complete deployment workflow:

- **Instances**: Short-lived environments for agents (IP-based access, no persistence)
- **Services**: Long-running services for apps (domain access, optional storage, multiple replicas)

## Prerequisites

Install AEnvironment CLI:

```bash
pip install aenvironment --upgrade
aenv --help
```

## Deployment Workflows

### Workflow A: Deploy with Local Image Build

Build Docker image locally, register to EnvHub, and deploy.

**When to use**: Creating a new environment from scratch with local Dockerfile.

**Requirements**: Docker installed, registry credentials configured.

**Script**: `scripts/deploy_with_local_build.py`

**Example**:

```bash
python scripts/deploy_with_local_build.py \
  --env-name myagent \
  --owner-name john \
  --api-service-url https://api.example.com \
  --envhub-url https://envhub.example.com \

Report a problem

Foxx AIAEnvironment Deploy
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.