Plugins

Vision provider route that transcribes attached images to text through a configurable model (15+ OpenAI-compatible and Anthropic vendors) while DeepSeek keeps answering.

Reuses DeepSeek web's built-in vision mode for text-only models: the deepseek_vision tool drives the local deepseek-vision-cli browser automation (manual login helper, deep-think enabled, auto-closes browser) and returns image descriptions as text.

Vision toolkit for text-only DeepSeek: model-invokable `vision` tool, wrapper adapters for deepseek/opencode-go (v4 flash/pro), Antigravity IDE quota (default, flash/pro) / any OpenAI-compatible VLM / Gemini / local Ollama channels, evidence memory with compaction rehydration, content-hash cache, and a bilingual client panel.

Model-facing `vision` tool for DeepSeek Harness: describe and OCR image files by calling the free Zhipu GLM vision API directly (glm-4v-flash fallback chain), no external CLI required.

Ingestion-time tool-output curation: keeps error and warning lines plus head and tail, spills the full text to disk with a pointer the model can read back; covers every text tool, small results pass through untouched.

Bi-temporal memory backed by a SodaMem daemon: recall is contributed to the prompt while each turn is assembled and every closed turn is ingested, with no tool call and no model call on the read path; each recalled fact cites the turn it came from and carries validity dates, so superseded facts stop being retrieved.

Real-time DeepSeek API usage: account balance with low-balance alerts (browser notification on threshold crossing), token & request stats (total/today/last-60s windows, per-model and per-session), estimated cost — a composer dock plus a settings-page usage panel.

Usage statistics dashboard for the DSH Web UI: info cards, a 280-day GitHub-style activity heatmap, daily token trend, and model-usage donut chart.

MiniMax token-plan quota dock for the Web UI: a drop-shape 5h-usage % chip in the conversation input area, click for a detail panel listing every model's 5h & weekly usage and reset countdown, auto-hidden when the active default model is not a MiniMax model.

DeepSeek API usage dashboard for DSH Web: daily cache-hit/miss and output token statistics from session logs, per-model cost estimates, account balance, 7-day trends, and a composer usage summary.

Declutter the DSH settings nav: fold third-party plugin entries under collapsible group rows, with bookmark-style custom groups, a fold toggle, and auto classification (market tags → name rules → your AI model).

File-browser web app for DSH: React frontend and REST API served by the host webserver, a /dsh-agfs command that opens at the current session workspace, and a browse_files model tool.

Subscription OAuth login: use ChatGPT/Claude/Grok/Kimi subscription accounts (not API keys) with automatic model discovery.

Incremental file mounting with read dedupe: mounted line ranges are never re-sent to the model, on-disk changes invalidate and remount, with a Mounted Files tab and token-savings accounting.

DeepSeek Harness (DSH) plugin: a "Thinking Language" selector under Settings → General that tells the model which language to reason in (chain of thought) via a system-prompt section. Zero extra calls, zero latency, 12 languages.

Expose DeepSeek Harness agent capabilities as an MCP server, so an external MCP client (e.g. Hermes) can drive Harness to execute coding tasks. Hermes = brain, Harness = arms.

Command Code Go provider adapter for dsh: scan the Go-plan model catalog dynamically and serve /alpha/generate through the harness LLM seam

Lightweight usage analytics for DeepSeek Harness — token trends, activity heatmaps, model breakdowns, and exports.

Replay-aware per-turn model usage and session cost for DeepSeek Harness

A model-native meta-agent plugin for DeepSeek Harness: the model synthesizes task-specific workflows at runtime from five agentic workflow patterns (prompt chaining, parallel workers, router, supervisor, evaluation loop) and coordinates tools and subagent

Per-model token cost tracking with configurable cache-hit/miss, output and peak-window pricing, a live session cost bar, and unconfigured-model flags.

Adaptive model routing: per-request complexity classification with automatic provider routing.

Claude-Code-style lifecycle hooks: configured shell commands run before/after model tool calls with a JSON payload on stdin; a non-zero pre-tool exit blocks the call.

First-class Git tools for the model: status / diff / log / commit / branch with validated messages and paths - no bare-shell git calls.