Plugins

A fixed-corner whale widget for the DSH web GUI — balance, today's usage and per-turn cost with peak/off-peak pricing, editable balance-alert and daily-budget bubbles, a module-based custom bubble queue with A/B weighted choices and random lines or images, 30+ vendor templates (OpenAI, OpenRouter, Kimi, SiliconFlow, Ark, Zhipu, MiniMax and more) with per-model balance and subscription quota, plus task-end sound, imported audio, custom roles and a resource manager. Local-only, no telemetry.

Per-session and daily API cost, budget with usage %, official balance, history dashboard, and one-click official price sync with peak/off-peak pricing.

Deep-read books and articles in five modes (quick, deep, knowledge map, Feynman, full book) with claim-evidence-data reports, Mermaid/XMind mindmaps, batch comparison, budget preflight, transparent background-job progress, and MD/HTML exports from URLs, files, or pasted text.

Floating DSH desktop pet with 12 authored state animations for requests, tools, compaction, archive and prompt enhancement; live context occupancy; a persistent cross-session Lifetime Ledger; per-provider/model totals, hourly trends and 7/30-day views; dark/light theme switching; optional cost estimates (off by default) with a visual price editor and monthly budget alert; client-side skin ZIP import; JSON/CSV usage export; snapshot-only panel reads with background incremental indexing; and opt-in prompt enhancement with editable preview and DSH composer submission.

Off-peak batch runner for DeepSeek Harness: users define tasks during busy hours — prompt, strategy, priority, workspace, locked files — and release them by hand; the plugin then runs the batch automatically inside user-defined off-peak windows. Each run passes five preflight checks (workspace, git HEAD, locked-file sha256, window fit, daily budget), supports four execution strategies (single / iterative / sampling / review) and three autonomy levels (L1 per-task, L2 batch, L3 full-auto), and closes the loop with a morning report containing diffs, actual spend and savings from a daily ledger aligned with official peak/valley pricing.

Match your models and token budgets to each role, then let DSH delegate specialized subagents by task type. Covers architecture analysis, UI/UX, code implementation, codebase exploration, and documentation research, with automatic setup plus per-role reasoning effort and tool permissions.

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

Local-first four-bucket Token observability: persistent session/provider/model/day dashboards, trends, budget and anomaly signals, public-rate estimates, safe aggregate exports, and explicit opt-in usage or session-trajectory analysis.

China A-share financial data MCP server for the harness: 11 fin_data__* tools (quotes/K-lines/financials/calendar/special-data/announcements/EDB/reconcile/funds/indices) over THS free official REST (primary) + AKShare (fallback) + Wind (exclusive domains), with observable degradation, context-budget compression (result-side -72.2% measured), and an offline 10-year market-data lake; BYOK.

Role-based multi-agent team for DeepSeek Harness. The /team command assembles up to 12 specialist subagents (product, architect, researcher, UI, backend, frontend, database, security, reviewer, QA, DevOps, docs) and runs a nine-phase gated pipeline (clarify, research, design, spec review, plan approval, implement, review, test, deliver) over a shared workspace of artifacts (SPEC, PLAN, TASKS, STATE, REVIEW, TEST). Quality gates are enforced by plugin code rather than requested in prompts: an unfinished task cannot be marked completed, quality findings must be adjudicated by the reviewer or QA role, and coverage gaps or rework over budget appear live in the overlay and in /team status. Ships a live team overlay (phases, roster with per-role models, task DAG, artifact preview, decision buttons), a persistent-team mode that resumes across sessions, and per-run token and time accounting. Zero runtime dependencies, no build step.

Per-plugin usage budgets and cost caps on the llm/stream waterfall: per-model usage accounting, session and monthly budgets with warn/block transitions, latency windows, and a carbon-footprint estimate.

Low-cost video understanding tool: a video_understand tool turns a Bilibili link / BV / local video into an AVIS info layer (ASR + scene structure + object tracks + YOLO labels) and returns summary + Q&A. Question-driven dynamic routing across layers (L0 ASR / L1 object tracks / L2 key-frame VLM), semantic-layer reuse for repeat questions, per-question budget cap. Python engine: core layer needs faster-whisper / opencv / yt-dlp (~200-300MB); the optional semantic layer adds ~2GB of torch / transformers / ultralytics. A bundled doctor --fix sets up the venv and installs both.

China A-share financial data MCP server for the harness: 11 fin_data__* tools (quotes/K-lines/financials/calendar/special-data/announcements/EDB/reconcile/funds/indices) over THS free official REST (primary) + AKShare (fallback) + Wind (exclusive domains), with observable degradation, context-budget compression (result-side -72.2% measured), and an offline 10-year market-data lake; BYOK.

Reads the files that AGENTS.md references with @path imports and injects their content into model context at session start, with byte budgets, compaction re-seeding and a six-language settings card.

Deep research as an agent mode: one reviewed loop that plans, researches in adaptive parallel rounds driven by the gaps the researchers themselves declare, synthesises a cited report and then reviews it adversarially, instead of a script the model rewrites per call. Questions the round budget never reached are named in the report rather than dropped.

Local-first usage and cost meter for the DSH web UI: composer button with live session cost and balance; panel with a stacked per-model daily/7-day chart priced per effective list-price interval (official pricing page synced daily), monthly usage tiers with a canvas share card, 30/50/85% context-pressure warnings with one-click /compact, budget alerts, and an in-panel npm update check with one-click update.

Cost tracking priced from an online catalogue (models.dev + OpenRouter, 8000+ models, each call priced at its own instant): a per-turn cost line, a Cost tab attributing spend to tool output / model output / system prompt / commands, budget, monthly forecast, ~166 live-rate currencies, and a bill_stats agent tool.

Governed long-term memory for DeepSeek Harness: append-only evidence ledger with write guards, 7-tier authority claims, supersede chains, and budgeted pre-step injection with per-domain quotas.

Long-term cross-session markdown memory with an LLM-Wiki structure and a SOUL.md persona, injected at session start (by default only after the first user message and only in the active session), plus remember/recall/consolidate/forget workflows, an embedded memory skill, pack/unpack migration tooling, and an idle-time first-person recall nudge that surfaces real memories in a companion voice. Also gives the store observability and upkeep it lacked: a per-file boot-block budget that keeps the index routing table from being truncated, automatic last_access stamping so salience decay actually works, a read-only memory health dashboard (size, freshness, lint-parity checks), a declarative index compiler (pages declare summary:, one command re-lines the index), an interactive memory graph (page links plus shared tags, draggable, wheel zoom), structured queries (tag/type/salience/hot/stale filters), cross-link suggestions for pages that have no relationship yet, and a `checkup` report that ends in a prioritised action list. The boot payload was also cut by ~23% by moving operational detail into the on-demand skill.

Four-module session companion: smart conversation export (Markdown/PDF/JSON/PNG long-image with privacy redaction and batch ZIP), context handoff summaries with template save/import, API cost optimization (live official pricing, peak/off-peak scheduling, daily/monthly budgets, model routing), and global conversation search with in-chat find (Ctrl+F, CSS Custom Highlight API).

A corner HUD for DeepSeek Harness that shows your session's token pressure, per-model cost, and daily/monthly usage at a glance, with an editable budget and balance that tracks spending for you. 右下角常驻的 Token 仪表盘：实时看会话压力、按模型估算花费。预算和余额点一下就能改，每天每月用了多少都有记录。

Usage and cost view for DeepSeek Harness: aggregates real token usage from existing session logs, locks each cost at the price of its event time, and shows month and today spend against a budget from a sidebar card or an input-bar pill.

Advisor reviews that converge over rounds, an engineering mode whose writes require a design token, escalation of a task to a stronger model, and parallel multi-model consultation; a settings page edits the model routes, timeouts and budgets.

Compaction fix for NInfer-hosted qwen3.8-27b gateways: disables thinking on dsh context-compaction and session-title calls so xhigh reasoning stops burning the whole output budget; the same idea applies to other launch methods, this package targets NInfer only.