Skip to main content

Installation

Agent Actions is available on PyPI and can be installed with pip, pipx, or uv. Let's get you set up to run your first agentic workflow.

Requirements​

  • Python 3.11 or higher
  • At least one LLM provider API key (OpenAI, Anthropic, etc.)

Quick Install​

uv pip install agent-actions

Using pip​

pip install agent-actions

Using pipx (for CLI tools)​

pipx install agent-actions

Verify Installation​

agac --version

You should see output like:

agent-actions 2.0.0

Provider Configuration​

Set API keys as environment variables or in a .env file.

Environment Variables​

# OpenAI
export OPENAI_API_KEY="sk-..."

# Anthropic
export ANTHROPIC_API_KEY="sk-ant-..."

# Google Gemini
export GEMINI_API_KEY="..."

Using a .env File​

Create a .env file in your project directory:

OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
GEMINI_API_KEY=...

Agent Actions automatically loads .env files from the current directory.

Supported Providers​

ProviderEnvironment VariableModels
OpenAIOPENAI_API_KEYAny model supported by the OpenAI API
AnthropicANTHROPIC_API_KEYAny model supported by the Anthropic API
GoogleGEMINI_API_KEYAny model supported by the Gemini API
GroqGROQ_API_KEYAny model supported by the Groq API
MistralMISTRAL_API_KEYAny model supported by the Mistral API
CohereCOHERE_API_KEYAny model supported by the Cohere API
Ollama(local)Any model you've pulled locally

Local Models with Ollama​

What if you want to run models locally without API keys? Ollama lets you run open-source models on your own hardware. This is useful for development, privacy-sensitive workloads, or when you want to avoid API costs entirely. Note that local models may have different capabilities than cloud providers—test your agentic workflow with your target model before deploying.

# Install Ollama
curl -fsSL https://ollama.ai/install.sh | sh

# Pull a model
ollama pull llama3.1

# Use in your agentic workflow
# vendor: ollama
# model: llama3.1

Upgrading​

uv pip install --upgrade agent-actions

# Or with pip
pip install --upgrade agent-actions

Development Installation​

For contributing to Agent Actions:

git clone https://github.com/Muizzkolapo/agent-actions.git
cd agent-actions

# Install with dev dependencies (recommended)
uv sync --dev

# Or with pip
pip install -e ".[dev]"

Next Steps​

Now that you have Agent Actions installed, let's build something: