A command-line tool for capturing screenshots and managing knowledge in Obris, your personal knowledge layer for AI. Save content to organized topics so you never start another AI chat from zero again. Requires Python 3.10+. Or run directly with uv: For desktop notifications with deep linking to your uploaded content: Generate an API key from your Obris dashboard. Don't have an account? Sign up. T
Add this skill
npx mdskills install obris-dev/obris-cliWell-documented CLI tool for screenshot capture and knowledge management with clear setup and usage examples
A command-line tool for capturing screenshots and managing knowledge in Obris, your personal knowledge layer for AI. Save content to organized topics so you never start another AI chat from zero again.
Requires Python 3.10+.
pip install obris-cli
Or run directly with uv:
uvx --from obris-cli obris --help
For desktop notifications with deep linking to your uploaded content:
brew install terminal-notifier
Generate an API key from your Obris dashboard. Don't have an account? Sign up.
obris auth --key
This saves your key locally to ~/.obris/config.json and detects your Scratch topic (the default destination for captures).
You can configure separate API keys for prod and dev side by side:
obris auth --key # saves to prod (default)
obris --env dev auth --key # saves to dev
Use --env on any command to override the active environment:
obris --env dev capture
obris --env dev topic list
Set the default environment:
obris env dev # switch default to dev
obris env # show current default
obris capture # screenshot + upload to Scratch
obris capture --name "my pic" # explicit name
obris capture --prompt # prompt for a name via dialog
obris capture --topic # upload to a specific topic
obris upload photo.png # upload to Scratch
obris upload photo.png --topic # upload to a specific topic
obris topic list # list all topics
obris topic list # list knowledge in a topic
obris knowledge move --topic # move to another topic
obris knowledge delete # delete a knowledge item
Bind keyboard shortcuts to capture commands using any automation tool — Alfred, Raycast, Keyboard Maestro, macOS Shortcuts, etc.
Create a workflow with a Hotkey trigger connected to a Run Script action (language: /bin/zsh):
Quick capture:
/full/path/to/obris capture
Capture with name prompt:
/full/path/to/obris capture --prompt
Tip: Use the full path to the
obrisbinary since Alfred doesn't load your shell profile. Runwhich obristo find it.
| Platform | Capture | Upload / Topics / Knowledge |
|---|---|---|
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| Windows | No | Yes |
git clone https://github.com/obris-dev/obris-cli.git
cd obris-cli
uv sync
Run commands locally without installing:
uv run obris --env dev auth --key
uv run obris capture
uv run obris topic list
make publish # build and publish to PyPI
This CLI sends your Obris API key to the Obris API (api.obris.ai) to authenticate requests. It uploads files and retrieves topic and knowledge data from your account. No data is stored beyond the local config file at ~/.obris/config.json.
For the full privacy policy, see obris.ai/privacy.
For issues or questions, contact support@obris.ai or open an issue on GitHub.
MIT
Install via CLI
npx mdskills install obris-dev/obris-cliObris CLI is a free, open-source AI agent skill. A command-line tool for capturing screenshots and managing knowledge in Obris, your personal knowledge layer for AI. Save content to organized topics so you never start another AI chat from zero again. Requires Python 3.10+. Or run directly with uv: For desktop notifications with deep linking to your uploaded content: Generate an API key from your Obris dashboard. Don't have an account? Sign up. T
Install Obris CLI with a single command:
npx mdskills install obris-dev/obris-cliThis downloads the skill files into your project and your AI agent picks them up automatically.
Obris CLI works with Claude Code, Claude Desktop, Cursor, Vscode Copilot, Windsurf, Continue Dev, Codex, Gemini Cli, Amp, Roo Code, Goose, Opencode, Trae, Qodo, Command Code. Skills use the open SKILL.md format which is compatible with any AI coding agent that reads markdown instructions.