swival 0.1.2

A small, powerful CLI coding agent for open AI models

Swival

A small, powerful, open-source CLI coding agent that works with open models.

Swival connects to LM Studio or HuggingFace Inference API, sends your task, and runs an autonomous tool-calling loop until it produces an answer. No configuration needed.

It auto-discovers your loaded model, gives it sandboxed file access, and gets out of the way.

It's what I use every day. Try it and see if you like it too.

Quickstart

1. Install LM Studio and load a model with tool-calling support (I recommend qwen3-coder-next). Crank the context size as high as your hardware allows.
2. Start the LM Studio server.
3. Install Swival:

uv tool install swival

4. Run:

swival "Refactor the error handling in src/api.py"

That's it. Swival finds the model, connects, and goes to work.

For interactive sessions:

swival --repl

To update to the latest version:

uv tool upgrade swival

To uninstall:

uv tool uninstall swival

Documentation

• Getting Started -- installation, first run, what happens under the hood
• Usage -- one-shot mode, REPL mode, CLI flags, piping, exit codes
• Tools -- what the agent can do: file ops, search, editing, web fetching, thinking, command execution
• Safety and Sandboxing -- path resolution, symlink protection, command whitelisting, YOLO mode
• Skills -- creating and using SKILL.md-based agent skills
• Customization -- project instructions, system prompt overrides, tuning parameters
• Providers -- LM Studio and HuggingFace configuration
• Reports -- JSON reports for benchmarking and evaluation
• Using Swival with AgentFS -- copy-on-write filesystem sandboxing for safe agent runs