lightroom-py
Run the following command in DeepSeek Harness:
dsh plugin install drshy-org/lightroom-py
Paste the following prompt into your AI chat to install this plugin:
Install the plugin in DeepSeek Harness by running dsh plugin install drshy-org/lightroom-py (source: https://github.com/drshy-org/lightroom-py).
About this plugin
Lightroom Classic has long been the hardest RAW editor to automate: no AppleScript, no COM, no UXP. The Lua plugin SDK is the only door, and the plugin side can only send outbound requests-it cannot host a server. lightroom-py closes that gap with a lightweight local HTTP bridge: a tiny .lrplugin polls a Python-side service, and Python or Claude drives the entire Lightroom UI through 62 bridge handlers and 80 CLI verbs.
Core capabilities span the full Develop pipeline-global settings, HSL, tone curves, lens correction, snapshots-plus programmatic radial and linear mask creation (radial masks verified with a 35.4 percent pixel-diff), triggering Adobe's AI adaptive mask compute path, hierarchical keyword and IPTC metadata writes, multi-format exports with watermarking and sharpening, smart collection management, and external-edit roundtrips. A SQLite read fast-path keeps even 50,000-photo catalogs instant to query.
Built for photographers who want Lightroom in an automation pipeline, batch-processing script authors, and anyone who wants Claude or any MCP client to operate Lightroom directly. One setup command installs the plugin, registers the macOS login LaunchAgent, and ships the agent skill-no manual token paste required.
Use Cases
- Apply a preset to a selection and export watermarked JPEGs via Claude
- Create radial masks programmatically for localized color grading
- Trigger AI adaptive masks and batch-process an entire photo set
Best For
- Photographers who want Lightroom in an automation pipeline
- Developers writing batch RAW-processing scripts
- AI agent users who want Claude or MCP clients to operate Lightroom directly
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.