Installation¶
Complete installation instructions for Inkwell.
Requirements¶
- Python 3.10+ - Check with
python --version - ffmpeg - Required for audio processing
- uv - Modern Python package installer (install uv)
Install Inkwell¶
Using uv (Recommended)¶
# Install uv if you haven't already
curl -LsSf https://astral.sh/uv/install.sh | sh
# Install inkwell as a tool
uv tool install inkwell-cli
Using pip¶
From Source¶
# Clone the repository
git clone https://github.com/chekos/inkwell-cli.git
cd inkwell-cli
# Install with uv (recommended)
uv sync --dev
# Or with pip
pip install -e .
Verify Installation¶
Install ffmpeg¶
ffmpeg is required for audio processing.
macOS¶
Ubuntu/Debian¶
Windows¶
Download from ffmpeg.org or use:
Verify ffmpeg¶
API Keys¶
Inkwell requires API keys for transcription and extraction.
Google AI (Required)¶
Used for Gemini transcription and content extraction.
- Go to Google AI Studio
- Click "Create API Key"
- Copy your API key
Configure via CLI (Recommended):
This key is also used for Gemini extraction unless you set
extraction.gemini_api_key separately. The Gemini transcription model is
configurable at transcription.model_name; generated config files currently
default to gemini-2.5-flash.
Or via environment variable:
export GOOGLE_API_KEY="your-google-ai-api-key"
# Make permanent (add to ~/.bashrc or ~/.zshrc)
echo 'export GOOGLE_API_KEY="your-key"' >> ~/.zshrc
Anthropic (Optional)¶
Required only for Interview Mode.
- Go to Anthropic Console
- Create an API key
- Configure:
Configuration¶
Inkwell creates configuration files automatically on first run.
Default Locations¶
| Platform | Config Directory |
|---|---|
| Linux/macOS | ~/.config/inkwell/ |
| Windows | %APPDATA%\inkwell\ |
Files Created¶
~/.config/inkwell/
├── config.yaml # Global settings
├── feeds.yaml # Feed definitions
└── .keyfile # Encryption key (auto-generated)
Verify Setup¶
Distribution¶
| Method | Status |
|---|---|
uv tool install inkwell-cli |
Supported (recommended) |
PyPI package (pip install inkwell-cli) |
Supported |
| Docker image | Not yet supported |
| Homebrew formula | Not yet supported |
Docker and Homebrew packaging are not currently provided. If either install path would help your workflow, open a GitHub issue so demand can be tracked.
Next Steps¶
- Quick Start - Process your first episode
- Managing Feeds - Add your podcasts
- Configuration - Customize settings