Plugins

Per-request thinking budgets, a compaction backend, and a settings tab for the local Qwen3.8 line (llama.cpp or NInfer).

Embedded C/C++ firmware development toolbox: agents and skills covering FreeRTOS, interrupts, NVM storage, Keil MDK, ARMCLANG, HardFault triage, state machines, architecture, and LVGL patterns.

Forces context compression when running low-context llama.cpp models such as Qwen3.8-27B, improving model efficiency and usage smoothness, together with well-refined compaction prompts.

Bilibili video analysis: metadata, transcript (ASR fallback via Bijian/sherpa-onnx/whisper.cpp), comments, danmaku, and sharp keyframes with optional local vision descriptions.

Voice input for the web UI: dictation chunked by pauses and voice messages, each with its own provider fallback chain (Deepgram, Groq, HuggingFace, local whisper.cpp, or an OpenAI-compatible endpoint).

Local-first hands-free AI voice assistant plugin for DeepSeek Harness (DSH), with speech-to-text, text-to-speech, voice commands, and local speech engines.

Local model plugin for DeepSeek Harness: manage GGUF models in settings, auto-launch llama.cpp on first message, and auto-unload after 5 minutes of idle time to free VRAM.

C/C++ LSP configuration plugin for DeepSeek Harness using clangd.

Fronts stdio MCP servers with three router tools (list_capabilities, describe_tool, call_tool) so schemas load on demand instead of riding every request; shapes every tool result through a deterministic output ladder that archives the original first (retrieve_spill reads it back); prices tool groups in tokens and can hide built-in groups per agent. Bridges to a standard-library Python router, so Python 3.9+ is required.

DeepSeek Harness (DSH) plugin for llama.cpp docker panel "llamapad" / llamapad 面板的 DeepSeek Harness 插件

Personal dsh preferences: a tunable companion to the official compaction engine (threshold, retention, summarization model/reasoning, turn-end compaction), llama.cpp model import for the Models page, and tab/notification helpers (status-light favicon, bro

OpenAI-compatible client for llama.cpp / Unsloth Desktop (default :8080).

Local-first voice conversations for DSH, with local speech recognition and synthesis and optional external providers.

Per-provider concurrency gate on the llm/stream waterfall: holds surplus model requests in a FIFO queue inside DSH so main agent, subagents and compaction take turns on a single-slot local server (llama.cpp --parallel 1) instead of timing out; maxConcurrent / maxQueued / queueTimeoutMs.

Host-side long-task detection for small-context local models: long outputs are scaffolded into a parts skeleton (one part per reply, then assemble), and multi-step work into a PLAN.md ledger with notes/ (plan first, one step per reply). Detection is heuristic and host-side, so it does not depend on the model judging its own limits.

C++ audit skill: detects the target project's actual C++ standard before making any recommendation, then reports findings across RAII and ownership, move semantics, exception safety, performance and memory, containers and algorithms, concurrency, templates and compile-time, standard boundary, and type safety, ABI and interface design, each with a checkpoint, a reason, a bad-to-good example and the minimum standard.

C++ runtime info probe for Windows via CDB: cpp_trace prints variables at multiple breakpoints in one session (replaces hand-written print), cpp_probe dumps the call stack at a conditional breakpoint. Install: dsh plugin --profile web add dsh-cpp-probe.

Re-encodes WebP attachment images to PNG so alpha-channel images load on LLM backends that cannot decode WebP (e.g. llama.cpp).

DeepSeek Harness (dsh) plugin: live local LLM endpoint health in the web rightbar. llama.cpp slots (busy/TTFT/wedged, /metrics) + Ollama MODELS (loaded vs empty). Engine fingerprinted by response shape. Installable dsh.bundle.

DSH plugin: inject per-model sampling parameters (top_p, top_k, min_p, repeat_penalty, presence_penalty, frequency_penalty) into every chat-completions request sent to a local llama.cpp / llama-server gateway.

Keeps every model request inside a per-provider allow-list, enforced at the agent/request waterfall: disallowed models are redirected to a fallback or refused; subagents stop inheriting stale creation-time routes; warns when consecutive requests would make a --models-max 1 llama.cpp router reload models. A one-entry list is a global single-model mode.

Control panel for local LLM inference servers in the DSH web view: service registry with health checks and stop controls, named parameter profiles for llama.cpp, SGLang and vLLM with pre-save KV/VRAM validation, GPU detection, and tok/s benchmarks.

Dictate prompts into the DeepSeek Harness Web Client — a microphone in the composer, with swappable transcription: hosted Whisper API, self-hosted server, or a fully offline whisper.cpp binary.