dsh-find-skill
Run the following command in DeepSeek Harness:
dsh plugin install Moximxxx/dsh-find-skill
Paste the following prompt into your AI chat to install this plugin:
Install the plugin into DeepSeek Harness using the command "dsh plugin install Moximxxx/dsh-find-skill" from the GitHub repository at https://github.com/Moximxxx/dsh-find-skill.
About this plugin
In AI agent development, LLMs often face limitations when preloaded tools are insufficient for complex tasks. Manually installing skills is time-consuming and inflexible, failing to meet dynamic demands. The dsh-find-skill plugin addresses this by seamlessly integrating the open vercel-labs/skills ecosystem into DeepSeek Harness. This plugin empowers LLMs with autonomous capabilities: when encountering tool gaps, they can automatically search for skills (skill_find), ask users for selection and install (skill_install), and manage removal (skill_remove). Skills can be loaded temporarily for the current session only or persisted to project/global scopes, with all operations performed in isolated environments to ensure safety and cleanliness. Ideal for developers using DeepSeek Harness to build AI workflows, dsh-find-skill simplifies skill lifecycle management, enabling LLMs to dynamically adapt to task changes and enhance agent automation and extensibility. In summary, through intelligent search and flexible installation, the plugin reduces manual intervention, allowing LLMs to focus more on task execution and driving the evolution of agent workflows.
Use Cases
- Automatically search for skills when LLM tasks exceed preinstalled tools.
- Dynamically install and manage skills during AI agent development.
- Switch skill scopes based on different project requirements.
Best For
- AI agent developers.
- Engineers building workflows with DeepSeek Harness.
- Users needing to dynamically extend LLM capabilities.
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.