Skip to main content

Python tool for setting up isolated AI environments

Project description

aicrate - containerize your AI agents

aicrate is a Python CLI tool that packages and orchestrates AI agents, skills, and MCP (Model Context Protocol) servers as OCI artifacts in isolated container environments. It enables you to build reproducible, shareable AI workspaces by bundling these artifacts and tools into unified pods.

Prerequisites

Installation

pip install aicrate

Basic Usage

# Run a containerized AI agent with the current directory as workspace
# and no additional skills or agents
$ aicrate run

# ...use a different workspace, and skills and agents as defined in the config
$ aicrate run --workspace ~/myproject --config ./examples/aicrate.yml

After running aicrate run, the user is automatically dropped into the /workspace.

Configuration

Define your AI environment in a YAML configuration file:

aicrate:
  image: quay.io/rhivos-ai-tools/aicrate:latest
  workspace: ~/
skills:
  - quay.io/aicrate/anthropic/claude-api:latest
  - quay.io/aicrate/anthropic/pdf:latest
  - quay.io/aicrate/obra/test-driven-development:latest
agents:
  - quay.io/aicrate/agency/engineering-code-reviewer:latest
mcp:
  - image: quay.io/aicrate/mcp-testing-farm-mcp:latest
    port: 8081
    env:
      - MCP_TRANSPORT: sse
      - MCP_PORT: 8081

See examples/aicrate.yml for a complete configuration with 30+ skills and agents.

See aicrate on quay.io for pre-built and ready to pull artifacts.

Similar projects

  • LobsterTrap/puzzlepod
    PuzzlePod and aicrate leverage Podman and systemd and run the agents inside a container. However, while PuzzlePod adds a governance layer on top (e.g. automated commit/rollback decisions), aicrate leverages OCI artifacts to manage skills and agents in OCI registries.
  • NVIDIA/OpenShell
    OpenShell provides a safe, private runtime for autonomous AI agents. Its goal are multi-tenant enterprise deployments while aicrate is intended for local usage by a single developer.
  • fletchgpc/agentbox
    agentbox provides a container-based environment for automated AI agents for more safety. aicrate adds the concepts of skills and agents as OCI artifacts to use, share and maintain them.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

aicrate-0.4.0-py3-none-any.whl (387.3 kB view details)

Uploaded Python 3

File details

Details for the file aicrate-0.4.0-py3-none-any.whl.

File metadata

  • Download URL: aicrate-0.4.0-py3-none-any.whl
  • Upload date:
  • Size: 387.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for aicrate-0.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 eeba684b98e5a8c1eebcb45aaffb43a20062b4f8df9cfeecb6d1fde0ab3dabdf
MD5 0b7f945df983796bc05db0be81fdb29c
BLAKE2b-256 4104ef4faf0232b8bb5e5cd6b47f3545348340d5d9e7c07ec33d0f8f903c9648

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page