CLI Changelog

​

Initial Release

​

Project lifecycle

• cirron init scaffolds projects. Can start from scratch or use PyTorch, TensorFlow, scikit-learn, or custom templates, with inference and training variants
• cirron compile, cirron build, and cirron deploy take a project from source to a deployed endpoint, driven by cirron.yaml
• cirron test runs the built-in suite for ML environments, data pipelines, and inference

​

Run and inspect

• cirron run pipeline | job | inference | sweep to launch work, with status, logs, and cancel for in-flight runs
• cirron replay re-executes a recorded run for reproducibility
• cirron list and cirron info summarize project and platform state

​

Local-first profiling

• cirron traces view | list | export | clear operates directly on the SDK’s ./.cirron/ directory to render a flamegraph in the terminal, list sessions, or export traces
• cirron spool inspect | flush | clear manages the on-disk spool and ships artifacts to the platform when you’re ready
• cirron traces snapshots and cirron traces snapshot surface weight and gradient snapshots from the spool

​

Plan, sync, and diagnose

• cirron plan compile | build | diff | save previews execution and persists plans for review
• cirron push, cirron pull, and cirron sync move artifacts between the local project and the platform
• cirron doctor, cirron diagnostics, cirron hardware, and cirron lint validate the environment before a run

​

Auth and config

• cirron auth login | logout | status | refresh for API keys and session state
• cirron config, cirron settings, and cirron env for global, project, and per-environment configuration

​

Install

• npm install -g @cirron/cli
• curl -fsSL https://cli.cirron.com | bash
• Standalone binaries for linux-x64, macos-x64, macos-arm64, and win-x64