agent-core

Common building blocks for creating AI agents.

Production-ready components for building AI agents, optimized across LLMs and providers. Uses InteropRouter as a unified AI model provider interface.

Note

This library is in early development and subject to change.

Getting Started

Installation

# With uv.
uv add agent-core-toolkit

# With pip.
pip install agent-core-toolkit

Usage

Run the agent CLI.

uv run agent-core --prompt "Your prompt here" --working-dir /path/to/directory

With all options:

uv run agent-core --prompt "Your prompt here" --working-dir "/path/to/directory" --mode "permissive" --model "gpt-5.1-codex-max" --model-friendly-name "gpt-5.1-codex-max" --model-knowledge-cutoff "Sep 30, 2024" --timezone "America/New_York"

Or run with uvx:

uvx --from /path/to/agent-core agent-core --prompt "Your prompt here" --working-dir "/path/to/directory"

Development

Prerequisites

• uv
• prek

Setup

Create uv virtual environment and install dependencies:

uv sync --frozen --all-extras --all-groups

Set up git hooks:

prek install

To update dependencies (updates the lock file):

uv sync --all-extras --all-groups

Run formatting, linting, type checking, and tests in one command:

uv run ruff format && uv run ruff check --fix && uv run ty check && uv run pytest

Further Information

docs/DEVELOPMENT.md