browser-agent

Terminal browser-control agent.

Install

python -m venv .venv
.\.venv\Scripts\activate          # Windows
source .venv/bin/activate         # Linux/macOS
pip install -e .

Use

browser-agent "find the price of bananas at trader joe's" --model ollama:llama3.2
browser-agent "search wikipedia for Alan Turing" --model openai:gpt-4o-mini
browser-agent "..." --model vllm:meta-llama/Llama-3.1-8B-Instruct --base-url http://localhost:8000/v1

LLM backends:

• openai:<model> — any OpenAI-compatible endpoint (uses OPENAI_BASE_URL / OPENAI_API_KEY)
• vllm:<model> — alias for openai backend; pair with --base-url
• ollama:<model> — local Ollama daemon

Test Run

browser-agent "go to huggingface, go to deepseek r1 model card and give me a description of the deepseek r1 model" --model openai:openai/gpt-oss-20b:free --base-url https://openrouter.ai/api/v1 --api-key <API_KEY> --no-vision