dsh-plugin-skills-laoboshi
Run the following command in DeepSeek Harness:
dsh plugin install Modole/dsh-plugin-skills-laoboshi
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install Modole/dsh-plugin-skills-laoboshi in your DeepSeek Harness terminal to install; source code is available at https://github.com/Modole/dsh-plugin-skills-laoboshi .
About this plugin
Before this plugin, project-level and user-level Agent Skills in DSH Web were typically managed through scattered config files, ad-hoc scripts, and manual editing, with no unified interface to view or adjust them. dsh-plugin-skills-laoboshi extracts the built-in Skills management logic from Laoboshi Agent Studio into a standalone plugin, giving developers a single place inside DSH Web to browse, edit, and orchestrate both project-level and user-level Agent Skills.
Once installed, the plugin reads the dsh.bundle manifest and attaches its functionality to the active profile automatically, with no extra environment variables or startup scripts required. It runs under the current user permissions: project-level Skills stay scoped to their project, while user-level Skills follow the account across projects. The two tiers are clearly separated in the UI, reducing the risk of accidental changes affecting unrelated workloads.
If you are running multiple projects in DSH Web, need to tune Agent skill sets frequently, or want your team to share a consistent Skills management experience, this plugin fills that gap in the DSH ecosystem. It is lightweight, released under the MIT license, and ready to use for individual developers and small teams alike.
Use Cases
- Browse and edit project-level Agent Skills directly in DSH Web
- Separate and manage user-level vs. project-level Skills configs
- Give a team a unified entry point for Agent Skills management
Best For
- Developers running multiple Agent projects in DSH Web
- Small teams that adjust Agent skill sets frequently
- Users migrating from Laoboshi Agent Studio to DSH
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.