plain-agent 0.1.0

A small agent that calls OpenAI-compatible APIs.

Plain Agent

A small agent that calls OpenAI compatible LLM APIs.

This project starts with a streaming agent loop:

Interactive terminal
  -> reads your prompt
  -> streams assistant text as it arrives
  -> detects tool calls from the model
  -> runs workspace tools when requested
  -> sends tool results back to the model
  -> repeats until the assistant gives a final answer

Install

This project uses uv to track the Python environment. If uv is not installed, follow the official installation guide.

uv sync

Configuration

Create a local .env file or export environment variables in your shell. See .env.example for more examples.

For DeepSeek:

export DEEPSEEK_API_KEY="your-api-key"
export LLM_PROVIDER="deepseek"
export LLM_MODEL="deepseek-v4-flash"

For OpenAI:

export OPENAI_API_KEY="your-api-key"
export LLM_PROVIDER="openai"
export LLM_MODEL="gpt-5.4-mini"

You can still set LLM_BASE_URL when you want to override the provider default, such as pointing at a local OpenAI compatible server like Ollama.

Run

uv run plain-agent