Плагины

Pushes DSH approval and question cards to IM channels (WeChat first) and answers them from chat, with a session console, per-channel security, and an open push service.

Standalone mobile surface for dsh: scan-to-pair device trust, realtime SSE sync, three-channel completion notifications (in-app / Web Push / Server酱·Bark·Telegram), offline outbox, approval & question panels, per-provider usage meters, PWA; a native narrow-screen phone UI.

Audits every plugin HTTP route in a profile for a browser-trust fence: plugin routes win the web server's longest-prefix match ahead of the /api gateway, so they never see its trust check and must pin the Host to loopback themselves. Grades PASS/WARN/FAIL per route and fails a fence that compares Origin to Host without pinning it (bypassable by DNS rebinding). Ships as a CLI for CI and a route_fence_scan tool.

Tools guard that moves agent-issued `rm` targets to the macOS Trash instead of deleting, with a shell-aware lexer covering compound and disguised commands; a switch in Settings → General turns it off.

Background security auditor for DeepSeek Harness: scans agent outputs for secret leakage, checks command safety before execution, and surfaces findings in a persistent audit log.

Crash protection for installed plugins: crash cards with one-click disable, static pre-flight inspection, black-box crash telemetry, and a safe-mode rescue profile for a broken UI.

Auto-adjust model reasoning effort per task (off/low/high/max): task classification, \[effort=...] markers, in-turn error escalation, DeepSeek peak-pricing-period capping with countdown notices, and manual-selection priority.

WSL (Windows Subsystem for Linux) bridge for DeepSeek Harness: run Linux commands, manage distros, convert paths, and copy across the Windows/Linux boundary.

Microsoft Agent Framework (MAF) plugin for DeepSeek Harness: validate, run, and scaffold declarative agents and workflows (YAML).

Real AST-based scan of installed plugins for malicious-intent patterns (dynamic eval, cross-plugin writes, exfiltration-shaped network calls), with an optional periodic schedule and auto-quarantine on critical findings.

Tiered task queue as the single source of truth for DSH↔Hermes work (reset > approve > review): enqueue tools for all submission paths, lease/claim model with concurrency 1, priority aging against starvation, atomic busy mutex, and a Hermes-side consumer (cron */1 + monitor gating) that drains tasks, writes review-handoff requests with single-slot checks, and invokes the reset agent.

Hands DSH reset requests to an external ops agent via a versioned JSON protocol: preflight snapshot, pre-restart maturity gate (no pending approvals, free disk, cooldown), restart, health check, and interrupted-work recovery, with the result delivered back to the requesting session.

Auto-approval agent for DSH permission requests: rule layer (denyAlways/allowlist) plus Hermes Pro semantic adjudication with quick-fail retries, fail-closed to human when Hermes is unavailable, rejection feedback returned to the requesting session, and approve tasks enqueued into the task queue with a shared busy mutex.

Replaces the headless startup to add --model provider/model, --resume <id|latest>, --continue and --session-info flags to dsh headless runs.

Quality suite: background tsc --noEmit diagnostics fed back after the agent edits TypeScript, plus a /code-review command running five parallel review lenses with confidence scoring; absorbs dsh-code-check and dsh-command-code-review.

Evidence-first attention supervision for DeepSeek Harness with C0–C3 policy, deduplicated Inbox alerts, quiet hours, and redacted outcome records.

ServerChan3 (Server酱) push notifications when a turn finishes — title, model, project directory, git branch, status and a reply excerpt, with subagent filtering and fire-and-forget delivery.

Real Windows OS toasts for DeepSeek Harness: task-finished, needs-input and needs-approval alerts that persist in Notification Center under DeepSeek branding. Windows-only, zero runtime dependencies.

Reminder bubbles beside the input for DeepSeek Harness: any plugin or AI can push reminders (text/link) with chat-style stacking, drag-resize, AI summary, visual customization and random outfit presets. A static profile plugin available in every preset without disappearing on restart.

Task-complete and question notification sounds for the DeepSeek Harness Web UI: a soft chime when the agent finishes a turn, a ding-dong when it asks you something (questions, plan reviews, approvals). Synthesized Web Audio, zero assets, mute toggle and volume slider in the session header and settings.

One-click update for the local DeepSeek Harness checkout — compare versions, pull and rebuild, then restart, and roll back to any past release, all from a sidebar panel with live progress.

Git Changes tab for the DSH web GUI: staging, diffing, AI-drafted commit messages, branch/merge/stash/worktree operations and sync, with live updates via fs.watch plus a changeToken poll.

Task Runner mode for DeepSeek Harness: the main agent decomposes large tasks into independent subtasks, dispatches them to subagents (default local model) in fresh short contexts, and synthesizes results; concurrency and worker/orchestrator models are configurable in the settings panel.

Scheduled DeepSeek Harness Agent tasks with mandatory end-to-end tests, run history, and bidirectional delivery through nine messaging channels.