DeepSeek Harness Plugins · Model Inference
Plugins that extend DeepSeek Harness (DSH) with vision, workflows, memory, web tools and more.
Makes OpenCode Zen available as a local provider in dsh via an OpenAI-compatible bridge, no API key required, with opencode tools and MCP support.
A pure Node.js Fibonacci tool plugin with no network dependencies; install and invoke its registered tool in your session.
A pure Node.js array flattening tool with zero network dependencies, available as a registered tool in your session.
A pure Node-based Collatz sequence plugin with no network dependencies, ready to use in your session after installation.
A DeepSeek Harness model-inference plugin that humanizes Chinese AI output for more natural and fluent text.
A code review DeepSeek Harness plugin powered by model inference.
A pure Node.js absolute value plugin with no network dependency; invoke the registered tool in your session after installation.
A DeepSeek Harness plugin for database design in model inference workflows.
A vector search plugin for DeepSeek Harness that provides vector retrieval in model inference workloads.
Brings zero-latency streaming dictation, push-to-talk, and automatic multi-provider STT fallback to DeepSeek Harness, with seamless local or cloud transcription.
A DeepSeek Harness model-inference plugin that simplifies code via LLM reasoning.
Route OpenAI Codex models into DeepSeek Harness via ChatGPT OAuth—no API key needed, with device-code login and automatic token refresh.