Skip to main content

cmdbox

cmdbox is a command foundation for AI system development that lets you launch with minimal implementation. A single feature implementation can be exposed through CLI / REST API / MCP SV / Web UI / Edge UI / remote execution.

cmdbox operation image

Why cmdbox for AI System Developers

  • No repeated reimplementation
    • Implement a Feature command once, then use the same capability via CLI, Web UI, and REST API.
    • It is easy to promote a PoC CLI workflow into operational Web/API interfaces without rewriting business logic.
  • AI-ready components out of the box
    • Modes such as agent, llm, mcpsv, and a2asv let you manage agent execution, model operations, MCP exposure, and A2A integration on one platform.
    • You can separate tool execution, inference, knowledge retrieval, and external integration while keeping a consistent operational model.
  • Rich Web UI for daily operations
    • Beyond APIs and CLI, cmdbox provides operational screens for commandlets, agents, filer, limiter, audit, and user management.
    • Teams can run, monitor, and control workflows from a browser without building a separate admin console.
  • Designed for remote execution
    • Server-side commands can run through Redis, which fits node separation and scale-out architectures.
    • Heavy workloads can be offloaded to servers while keeping client components lightweight.
  • Fast, configuration-driven extensibility
    • features.yml defines command discovery, aliases, argument rules, and agent execution rules, separating implementation from operations policy.
    • You can adjust exposed commands and execution policies without code changes, which is useful for environment-specific control.
  • Built-in governance capabilities
    • You can combine audit logging, authentication/authorization (signin / oauth2 / saml / rule settings), and limiter controls.
    • It is easier to trace who executed which command under which conditions, enabling enterprise-grade operational control.

Problems It Solves Well

  • You need to launch internal AI tools quickly
    • Start with CLI features, then expose the same capabilities internally via Web and REST API.
  • Multiple teams/environments need one shared foundation
    • Run the same codebase across dev/stg/prod while switching execution policies through configuration.
  • You need safe integration between agents and existing systems
    • Combine MCP/A2A integration with authorization rules to control the scope of agent actions.
  • You need usage control and cost governance for AI operations
    • Use Limiter settings to enforce quotas/rate rules, prevent abuse, and keep usage and billing behavior predictable.
    • Define reusable plans and billing policies (period-based or metered) to align access control with commercial and internal chargeback models.

What You Can Build

  • Custom command development
    • Support complex option definitions, client/server execution, and Web execution.
    • Organize functionality by mode / cmd and standardize execution through a shared option model.
  • LLM integration
    • Chat, embedding, and configuration save/load/list workflows.
    • Manage model configurations and treat inference calls as operational commands.
  • Agent integration
    • Save/load agent settings, integrate MCP servers, and connect A2A servers.
    • Expand agent capabilities in phases while keeping clear boundaries on exposed tools.
  • Web UI operations
    • Use built-in screens for commandlets, agents, filer, limiter, audit, and user management.
    • Manage day-to-day operations from the browser while keeping the same backend command model.
  • Operational controls
    • Audit logs, authentication, command execution policy, and REST API controls.
    • Publish safely with permissions segmented by organization or user group and full traceability.
    • Configure plan and billing operations to support subscription-style access windows and metered charging models.

Typical Use Cases

  • Internal knowledge search combining document processing + LLM
  • Converting routine operations into commands and running them from Web UI with no-code interaction
  • Letting agents call internal tools safely within controlled boundaries
  • Running operational administration (filer, limiter, audit, and user control) from the built-in Web UI

Installation

uv pip install cmdbox[app]
cmdbox -v

Docker Environment Setup

  • Redis
cmdbox -m cmdbox -c redis_install
cmdbox -m cmdbox -c up -C redis
  • PostgreSQL (used by some commands)
cmdbox -m cmdbox -c pgsql_install
cmdbox -m cmdbox -c up -C pgsql
  • Full cmdbox suite
cmdbox -m cmdbox -c server_install
cmdbox -m cmdbox -c up -C cmdbox

Quick Start

cmdbox -m server -c start
cmdbox -m web -c start --signin_file .cmdbox/user_list.yml
cmdbox -m mcpsv -c start --signin_file .cmdbox/user_list.yml
cmdbox -m a2asv -c start --signin_file .cmdbox/user_list.yml

Tutorial

The tutorial content has moved to the following documentation:

Support and Sponsorship

If you find cmdbox useful and would like to support its development, consider becoming a sponsor!

GitHub Sponsors

👤 Individual Sponsorship

  • $5/month - Recognition in README and SPONSORS list

🏢 Corporate Sponsorship

  • $100/month (Bronze) - README and SPONSORS list recognition
  • $500/month (Silver) - Priority issue resolution
  • $1000/month (Gold) - Custom feature consultation

See SPONSORS.md for details.

Your sponsorship helps us continue developing and maintaining cmdbox!

Contributing

We welcome contributions! Please feel free to:

  • Report bugs via GitHub Issues
  • Submit pull requests with improvements
  • Share your use cases and feedback

License

This project is licensed under the MIT License, see the LICENSE file for details

Metadata

Release files for cmdbox 0.8.6.7

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for cmdbox 0.8.6.7
File Size Uploaded Provenance Actions
cmdbox-0.8.6.7.tar.gz 2.2 MB Without provenance Details for cmdbox-0.8.6.7.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for cmdbox 0.8.6.7
File Interpreter ABI Platform Actions
cmdbox-0.8.6.7-py3-none-any.whl Python 3 none any Details for cmdbox-0.8.6.7-py3-none-any.whl

Total release size: 5.1 MB

Release files / cmdbox-0.8.6.7.tar.gz

Download URL cmdbox-0.8.6.7.tar.gz
Size 2.2 MB
Tags Source
SHA-256 checksum
How to use checksums
f5e132ad7a4374e89af5c620270820966fd8346d038b62f7c87042cd41c776dd
BLAKE2b-256 checksum
How to use checksums
53e886b0c3ad94c2e8613ed4631e8698808ccf989e1d26ea2549fa8b669bf5de
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.17

Release files / cmdbox-0.8.6.7-py3-none-any.whl

Download URL cmdbox-0.8.6.7-py3-none-any.whl
Size 2.9 MB
Tags Python 3
SHA-256 checksum
How to use checksums
b9f53f60c21ca95345f5b51a73f2dcd6bb3dd76b572084c1d180291d3fe51c7c
BLAKE2b-256 checksum
How to use checksums
0904fd54d125e083091f7ef6e04eea657317267c7f9fe2d00b86e1c5bb551bd3
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.17

Security for cmdbox 0.8.6.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.8.6.7 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page