How to Use Pythinker Code, the Terminal-Native AI Engineering Agent

Learn how to install, configure, and use Pythinker Code for review-first coding workflows, debugging, ACP, MCP, local models, and privacy-conscious AI engineering.

Install Pythinker Code

macOS, Linux, WSL: curl -fsSL https://pythinker.com/install.sh | bash
Windows PowerShell: irm https://pythinker.com/install.ps1 | iex
Homebrew: brew install Pythoughts-labs/pythinker/pythinker-code
Python fallback: pip install pythinker-code

First run

Run pythinker --version, pythinker login, then pythinker from your repository.

Pythinker Code educational guide artwork

Practical workflows

Use Pythinker Code for review-first diff critique, security review, root-cause debugging, scoped implementation, repository understanding, and automation.

ACP, MCP, and local models

Run pythinker acp for editor integration, pythinker mcp for tool servers, and pythinker login --lm-studio or pythinker login --ollama for local models.

Privacy

Pythinker Code is the agent framework, not the LLM provider. Disable optional telemetry with pythinker --no-telemetry or PYTHINKER_DISABLE_TELEMETRY=1.