Plugins

Boot-time guard that repairs corrupt session logs, preflights plugin composition, vm-checks client artifacts, smoke-tests host apply() and quarantines crash-causing bundles, so a broken plugin can never brick startup.

System-native notifications when a task finishes or needs your confirmation: Windows toast, macOS osascript, and Linux notify-send.

Provider-neutral CI/CD capability for DeepSeek Harness with GitHub Actions support, bounded failure evidence, source-aware diagnostics, and approval-gated rerun/cancel controls.

Minimal prompt x full PTC tools: one-line persona, Code Mode SDK, subagents, workflows, web search, and Windows Git Bash.

A structured multi-panelist review workflow for the DeepSeek Harness: N panelists score in isolated subagents, blind to each other; a chair reconciles disagreement by anchor match then Δ-level adjudication; an independent critic audits the process. Works for design, code, paper or project review.

Calendar and hierarchical task workbench: AI-assisted intake clarification, consultation, decomposition, execution with user acceptance, and review, plus reminders and per-task AI session workspaces.

Registers the commercial-ui-ux skill for DSH: task-aware commercial UI/UX/GUI design, review, repair, and implementation (SaaS, dashboards, admin panels, forms, tables, design systems) with a reference-doc system and quality gates.

Contract-drafting agent preset: 11-stage lawyer workflow with 5-way parallel AI review, decision gates, and domain packs for general, employment and equity-investment contracts.

Task-to-skill pairing meta-skill with a laziness ladder for minimal loading, a user-editable routing table, and a quarantine, SkillSpector scan, human-approval install flow.

Mermaid diagram renderer for the Web UI conversation: renders mermaid code blocks into interactive cards (zoom, fit-width, fullscreen viewer, preview/code toggle) with official UI primitives, and keeps the original code block with an inline error banner when rendering fails.

Agent-initiated voice calls: `offer_call` rings the human (接听/拒接/稍后再说); accepted calls synthesize and play locally via CrispASR + Qwen3-TTS (9 speakers, 2 Chinese dialects), rejected calls return the decision to the agent.

Voice AI girlfriend for the Web UI: FunASR mic input, Qwen3-TTS spoken replies, companion animation window, and two-way QQ chat (text/voice/image push) via NapCat.

Speech-to-text voice input for the web UI: a mic button in the composer transcribes speech into the draft via the browser Web Speech API (zero-config) or an OpenAI-compatible Whisper API (OpenAI / Groq), with a selectable model and language in Settings.

Per-event sound notifications: turn completion, approval, question, plan-review, goal-blocked, and task-failure each get their own sound and volume, configurable in the Web UI (built-in synth, mute, or local audio file).

Chat image-attachment bridge with a `view_image` tool for any OpenAI-compatible VLM (local Ollama or cloud): pasted/dropped images become `view_image` path markers before reaching text-only DeepSeek models.

`describe_image` tool: a vision bridge that sends images to mimo-v2.5 through the opencode Zen API (credential `OPENCODE_GO_API_KEY`, free route first with paid fallback) and returns text descriptions for text-only models, with native passthrough and ImageMagick transcoding of SVG/TIFF/HEIC formats.

Model-facing image_describe (识图) tool over the DashScope OpenAI-compatible API (qwen3.7-flash), plus a paste bridge: on text-only sessions, pasted images auto-convert to file paths at send time and render back in the transcript, so they never trip image admission. Bring your own DASHSCOPE_API_KEY; endpoint/model/budgets configurable, redirect-proof HTTP client, works in every agent preset.

Agent-callable vision tool that describes local images via any OpenAI-compatible vision endpoint you configure, with an optional multi-model cross-check and no built-in keys.

Auto-discovery vision bridge for text-only DeepSeek Harness agents: automatically finds an image-capable model from your configured providers and returns picture descriptions as plain text via a vision tool.

On-demand vision for text-only DeepSeek models: upload images, and the model calls a view_image tool backed by any OpenAI-compatible vision endpoint (Qwen/DashScope by default).

Local-first structured vision for text-only agents: images go to a local OpenAI-compatible VLM and come back as JSON evidence (summary, verbatim OCR, layout regions, entities/relations, colors, explicit uncertainty), with anti-hallucination fallback and an optional paste/upload bridge; zero cloud cost, images never leave the machine.

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.

Routes chat images to a fixed OpenAI-compatible vision model, returns factual observations to the selected main model, and reuses session-scoped observations across replay, compaction, and restarts.