AI Agent Hub
Back to plugins
🧰

dsh-plugin-search

Web Tools Updated 2026.08.20

Run the following command in DeepSeek Harness:

dsh plugin install zoahdev/dsh-plugin-search

Paste the following prompt into your AI chat to install this plugin:

Please install this plugin in DeepSeek Harness via the command line, using the source code repository address https://github.com/zoahdev/dsh-plugin-search.

About this plugin

When building Agents with DeepSeek Harness, one of the most frustrating challenges is not knowing which tools are available in the plugin registry, often requiring developers to guess based on experience before writing code. To solve this "blind exploration" problem, dsh-plugin-search was created, directly integrating plugin discovery capabilities into the Agent itself and transforming the way plugin development is collaborated on.

The plugin provides three powerful tools to scan the plugin ecosystem comprehensively. dsh_search_plugins can simultaneously query npm packages and the awesome-dsh-plugin curated list, intelligently deduplicating and labeling versions and sources. dsh_plugin_lookup handles precise lookups, fetching npm metadata while falling back to a GitHub search for Git-only plugins. Additionally, dsh_awesome_top allows you to quickly browse the current curated list to get a snapshot of the entire ecosystem.

Whether you are a developer building complex workflows or a user who wants Agents to actively search for needed capabilities, this plugin is an excellent assistant. It eliminates the uncertainty of "guessing plugins," allowing Agents to proactively search and verify resources just like humans, thereby significantly improving development efficiency and the flexibility of tool combinations.

Use Cases

  • Need to find a dsh plugin that tracks GitHub releases.
  • Want the Agent to automatically search for available tools before building workflows.
  • Want to browse the awesome-dsh-plugin curated list to understand the ecosystem status.

Best For

  • Engineers developing DeepSeek Agents.
  • Plugin maintainers who want to reduce guessing and improve development efficiency.
  • Users needing to integrate complex toolchains but unsure of available resources.