dsh-plugin-forge
Run the following command in DeepSeek Harness:
dsh plugin install CAI-MH/dsh-plugin-forge
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install CAI-MH/dsh-plugin-forge in your terminal to install; source code is available at https://github.com/CAI-MH/dsh-plugin-forge
About this plugin
There is no single, reliable path for writing DSH local plugin bundles today. Format specs live in scattered docs, battle-tested lessons go unrecorded, and every new plugin starts from scratch. dsh-plugin-forge turns that ad-hoc process into a repeatable workflow: read the latest spec and pitfall library, scaffold a loadable bundle, implement business logic, validate and install in one step, then log every gotcha so the next iteration starts smarter.
Four tools cover the full loop. plugin_forge_guide serves the current dph format spec plus a living experience vault; plugin_forge_scaffold generates a ready-to-load bundle skeleton in the active workspace; plugin_forge_install validates and writes the bundle into the local web profile; plugin_forge_experience records and retrieves lessons across all workspaces, seeded automatically on first use so newcomers never start from a blank page.
Ideal for developers writing their first DSH local plugin, for teams that want to institutionalize plugin development guidelines, and for senior engineers who have already burned through the same pitfalls and want a lightweight, zero-config way to keep the knowledge flowing.
Use Cases
- Learning the dph format for the first time and needing a scaffold plus spec reference
- Finishing a plugin and wanting to validate and install it to the local web profile in one step
- Logging a gotcha after debugging so the next iteration avoids the same trap
Best For
- New developers writing their first DSH local plugin bundle
- Senior engineers who want to codify plugin development into a team standard
- Developers who create and debug DSH local plugins frequently in day-to-day work
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.