Plugins

Cumulative token usage tracking for DeepSeek Harness, with per-model and per-day breakdowns, a settings-page dashboard and an online-changeable data directory.

Local-first personal usage insights in Settings, with 1/7/30-day token heatmaps and a 30-day calendar, token composition, model mix, and skill-call status.

Aipx toolkit: skills plus a Hub Console that fronts every MCP server with ~4 meta tools — server pool health, per-tool enable/disable, a tool catalog, and a search playground showing exactly what mcp_search hands the model.

A Codex-style right side panel for the DSH Web UI — session artifacts browser (live file tree, scan include, preview, reveal) plus a side chat that forks the current session with a model picker, human-pacing scheduled runs and smart send-to-bottom scrolling.

Keeps a local model's context at a size the GPU handles well: measures prefill speed on every request and, before each step, warns or compacts early on a hard token ceiling, an observed time-to-first-token budget, or a predicted cold-prefill budget; /context-budget shows the numbers and the cost of compacting now.

AI-generated curiosity & trivia plugin for the dsh web GUI (Curio in the English UI): 「开始探奇」 generates a fresh batch of unexpected knowledge (excluding everything already seen), each item unfolds into two deeper layers plus similar knowledge points, and 「探奇清单」 keeps the full history (batch, items, layers, similar points) in browser storage. Reuses the host LLM channel with zero config; when no model is available, a DeepSeek API key can be pasted into the panel. Output language follows the DSH UI language (zh/en).

Community Codex provider for DeepSeek Harness with ChatGPT OAuth, Codex models, image input, quota display, per-session Fast and transport controls, and safe stream recovery.

Use local MLX-LM or MLX-VLM models in DeepSeek Harness through a loopback OpenAI-compatible provider, with optional DSH-owned server startup.

Token usage statistics plugin for DeepSeek Harness - today's token consumption by model, cache hit rate, 7-day trend chart, optional header badge, persistent storage with auto cleanup.

一个DSH插件，为自定义模型配置输入类型(input)、推理模式（reasoningEfforts）

Free anonymous OpenCode Zen models inside DeepSeek Harness (DSH): native LlmAdapter, no API key, no child process; optional Go sidecar included

dsh plugin: keep llm-pi-ai provider routes' model lists in step with the pi.dev gateway, written through the official settings seam

A green pixel-Ork mascot for DSH Web that masks model thinking and output as Waaaaaaagh! until clicked to reveal, plus a random-length input mask.

Provider-scoped HTTP/HTTPS forward proxy for DeepSeek Harness: configure independent proxy URLs and enable/disable switches per LLM provider, with environment-variable references for proxy credentials and model discovery routing.

Legacy code modernization workflow (preflight -> assess -> map -> transform) with a Cordis mount plugin, offline Python scripts and smoke tests.

Dual-MediaPlayer daemon ($think loop + $fx one-shots) with volume ducking, mood-aware sounds (smooth/struggle/deep_think triggered by consecutive success/failure streaks and prolonged agent runtime), 15 built-in PCM WAV synthesised from sine/square waveforms, per-tool sound mapping with 51 categorised tools, custom WAV/MP3 drop-in with auto-conversion, and a full settings UI with dark-mode fix.

Auto-switch the DeepSeek route to the vision model on demand: flash main session switches (A), pro keeps deep reasoning and delegates image reading to a vision subagent (B), subagents always switch, with fatal-failure fallback. No manual model switching.

Tiered compaction hub: replaces the single 80% compaction cliff with a configurable ladder (default 30/50/70/90) whose tiers escalate from gentle to maximum consolidation laws, folds a deterministic keyword index into every checkpoint so facts accumulate across cycles, and summarizes through a model chain — primary plus up to 3 fallbacks, each with its own reasoning effort. Adds a Compaction module sub-tab and a Settings card; toggling off always restores stock compaction.

Replaces the official Models settings page with provider lifecycle management, per-model reasoning effort tiers, retry overrides, drag reorder and upstream model sync.

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.

Manage multiple API keys per provider in Settings → API 密钥: keep a pool of keys with notes, switch the active key with one click, and the switch really takes effect on the next model call.

DeepSeek balance and per-session cost in the DSH web composer dock, with official peak/off-peak pricing (peak hours Mon-Fri BJT 9-12 / 14-18, weekends off-peak) and built-in deepseek-v4-flash-vision-exp vision model rates billed at flash price (images billed as tokens, max 384 tokens each).

Clean-mode for the Web GUI: hover any element and right-click to remove it; removals survive reloads and are restorable from the settings page.

Obsidian knowledge-base ingest workflow as a DeepSeek Harness plugin: a bundled skill provider + an obsidian_validate_vault model tool, all declarative and composable.