Back to directory

Awesome DeepSeek Harness Plugins

A community-maintained directory of over 13,000 plugins for the DeepSeek Harness (dsh) coding agent, collected automatically from GitHub, validated with static checks before listing, and searchable through a free public API or an in-agent plugin browser.

Awesome DeepSeek Harness Plugins

This project catalogs plugins for the DeepSeek Harness ecosystem, discovering candidates both from submitted pull requests and by automatically scanning GitHub for repositories tagged with the dsh-plugin topic, then reconciling the full set periodically so inactive repositories are not missed and removed repositories are only delisted after a full pass confirms it. Every candidate must pass a static check, reading the default branch's file tree to confirm required manifest fields and that referenced patch files actually exist, without installing dependencies or executing any repository code.

The catalog can be installed as a plugin inside dsh itself, adding a searchable, filterable store to the agent's settings that only offers installs backed by a published npm package or a structured GitHub source matching the plugin's declared identity. A free public API supports anonymous searches with rate limits, or a higher rate limit after creating an API key, and the underlying website, CLI and marketplace application code live in a separate sister repository that can be forked to self-host an independent plugin marketplace.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.