dsh-mindseye
MindsEye: model-driven vision tools, structured evidence, and exact cache for DeepSeek Harness
277 results
MindsEye: model-driven vision tools, structured evidence, and exact cache for DeepSeek Harness
Single-model disable gate: hides denied models from every discovery path, blocks dispatch with an in-protocol error chunk, and exposes a settings-section UI for the denylist
DSH client plugin: collapsible model-provider groups in the composer model picker — click a provider row to fold or unfold its models
DSH-native multi-AI group chat: the host (user) runs a chatroom of configurable AI agents with per-agent models, role cards, mute controls, and passive/active speaking modes — pure Node.js/Cordis, no Tauri/Rust
Provider-folded model selector for the DSH Web composer with catalog-driven reasoning effort choices.
Provider-first model selector for the DeepSeek Harness model seat: pick a provider, then a model from that provider — current model always shows Model · Provider; search and hardening (v0.3.2, verified against DSH 0.1.1-rc.2).
Take over the native DSH approval window and plan review card so a reject can carry a description the model will see.
Model pricing and capability board for DeepSeek Harness: compare per-1M-token prices, find the cheapest route, and see coding/agentic/vision tags for every LLM you can connect — in the DSH settings UI.
Peak / off-peak rate badges and switch countdown for DeepSeek Harness — per provider, per model, in the model selector and the composer tool row. Judges each provider by its own time zone and schedule instead of DeepSeek-only hours.
Dense translucent-glass Universal Palette for DeepSeek Harness Web — federates Commands, Models, Sessions and Conversation Hits with deterministic ranking and verified public-contract interoperability.
Mobile UI fixes for the DeepSeek Harness web profile: a restart row, input-modality switches, a narrow-screen drawer and settings tabs, a full-width model menu, a stuck-request network chip, a scrollable question region, a throttled reaction to streaming message storms, a guarded subagent-catalog refresh, click-to-open lineage chips, a narrow-header collector sheet, and a throttled session-list render under background streaming load.
DeepSeek Harness plugin: model sampling sliders (temperature / maxTokens) applied to every provider via the agent/request hook, with live settings persistence.
Plugin Clinic: read-only health checks for the installed DeepSeek Harness plugin set — loader health, dependency integrity, version compatibility, install-script risk, duplicates, and patch integrity, delivered as a model tool, a Web dashboard, and JSON reports.
Fold each DSH Web turn's thinking, tool calls and context into compact expandable process boxes cut at every piece of model text, and reuse the shipped turn-process disclosure row as the toggle whenever it exists.
Oh My DSH — tiered model routing for DeepSeek Harness: think/build tiers, an automatic vision tier for images, and image generation through an ordinary chat model (e.g. gpt-5.6-luna on opencode-go), all configured from an 'oh my dsh' settings tab. Works in the Web (Desktop) surface and in the terminal (TUI).
Jenkins CLI plugin (dual-face): multi-server config page, model tools, and a per-workspace Run-Jenkins-Job entry with environment tabs. Bilingual UI (zh/en), no hardcoded paths, publishable to npm / GitHub
Context-aware voice input for DeepSeek Harness with Web Speech, local SenseVoice transcription, model polish, editable Composer drafts, and user-controlled sending
Conversation undo for DeepSeek Harness: /undo <n> rewinds the model-visible surface and the Web UI collapses the undone turns.
Explicit model routing plus composer/Plan picker for DeepSeek Harness
DeepSeek Harness plugin for local GGUF models without Ollama
One-click web search-engine switcher beside the DeepSeek Harness model picker.
DeepSeek Harness token usage plugin: a GitHub-style heat map (day/week/month) of local session logs with granularity-linked per-model breakdown, cache hit rate, and an instant hover-today badge on the sidebar icon. Read-only — nothing leaves your machine.
Semantic, configurable chat titles for DeepSeek Harness with model selection and one-click regeneration
DeepSeek Harness 的增强模型选择器:单层菜单(搜索 + 分组)+ 底部内联推理强度(Effort)滑杆。