dsh-permission-gate
A DeepSeek Harness (DSH) plugin that adds an AI-adjudicated 'smart' approval mode: safe operations are auto-allowed via an isolated side-channel LLM judge, everything else falls back to the normal human approval popup.
124 results
A DeepSeek Harness (DSH) plugin that adds an AI-adjudicated 'smart' approval mode: safe operations are auto-allowed via an isolated side-channel LLM judge, everything else falls back to the normal human approval popup.
DSH 静默视觉增强:主模型照常选择,图片自动交给固定视觉模型后以隐藏上下文返回主模型。
DeepSeek Harness plugin: a third-party image-generation relay. Adds the 生图配置 card (Settings → Plugins → 插件配置) for configuring an OpenAI-compatible (/v1/images/generations + /v1/images/edits), DashScope (通义万相 async-task), Replicate or fal.ai image API; registers image_generate (文生图/图生图/局部重绘/指令编辑), imagegen_status, imagegen_test, imagegen_configure tools; installs a 生图 skill so the model auto-triggers on 生图/画图 mentions.
Tick-rail conversation navigator for the DeepSeek Harness web UI — a slim rail with one tick per message you sent, a peak-falloff hover effect, message previews, and click-to-jump.
LLM pre-review for sandbox-escalation approvals: an independent-context LLM gate answers sandbox escalation requests before they reach the user, falling back to the user on any failure.
Text-to-speech for DeepSeek Harness: speak agent replies in the Web UI with a provider fallback chain (OpenAI, ElevenLabs, Google, Azure, Groq, Deepgram, OpenRouter, Edge, Piper, eSpeak).
Whole-disk ctx.fileReferences provider for DeepSeek Harness: Everything-backed instant search with a bounded cross-platform fallback index, fuzzy matching, and regex queries
DSH Feishu self-built app integration: receives text messages over the official Feishu WebSocket long connection, routes them to a DSH agent session, streams safe assistant text deltas through a Feishu CardKit card, and falls back to ordinary text when card streaming is unavailable. Acknowledges with an OK reaction (text fallback). Standalone Cordis bundle — no dsh source changes.
DSH host bundle plugin: builds a two-level (category + skill) recall index over all local SKILL.md files and routes a natural-language intent to the best-matching skills, with three-state (hit / low-confidence / miss) fallback.
DeepSeek Harness host plugin that mirrors agent state to a CodeBuddy StickS3 pet over USB CDC and routes physical A/B approval back into the approval waterfall.
微信风格皮肤 for DeepSeek Harness — WeChat-style light theme for DeepSeek Harness: green bubbles, WeChat green accents, conversation-list sidebar, a far-left function rail, message avatars, and a resizable composer. Light mode only; dark mode falls back to the official look.
Show code changes in the dsh web GUI conversation: inline line-level diff inside edit/write tool cards with a streaming reveal animation, plus a per-turn file-change summary card (git-first, argument-diff fallback).
Image generation for DeepSeek Harness: a generate_image tool with pluggable providers — the FAL queue API or any OpenAI-compatible images API. The picture is shown inline in the conversation; the model receives either a link (works with any chat model) or the image itself (needs dsh-vision-bridge or a vision-capable model).
DeepSeek Harness plugin: content-routing LLM adapter. Text-only requests go to DeepSeek; requests containing images go to a user-configurable OpenAI-compatible vision provider. Read_image works natively for text-only models.
Let text-only models see images in DeepSeek Harness: intercepts the llm/stream waterfall and transparently substitutes each image with a vision model's description.
Policy-only multi-source quota router for DeepSeek Harness: deterministic task profiles, ordered candidate chains, health-aware fallback, and observable decisions.
Erii (Sakura) themed boot splash animation for the DeepSeek Harness Web UI: falling sakura petals, a mint monster mascot with a cherry blossom on its head, a soft pink progress bar, and the line "Sakura, walk slower." Auto fades out after ~3 seconds; pure client-side, zero dependencies.
Multi-level provider fallback for same-named models in the DeepSeek Harness
Session topic titles for DeepSeek Harness: LLM-generated titles for sessions stuck on the prompt-excerpt fallback, retry on every turn end, plus a /topics-size endpoint for the hover-card size patch (out-of-tree plugin)
Control which tool schemas are injected into the DeepSeek Harness model context. Registers a tool_visibility manager tool AND a settings-page UI (Settings → 工具注入控制) with per-tool toggles; backed by the framework-native tools.restrict filter and persisted to a JSON state file.
Spend gate for DeepSeek Harness: refuses over-budget model calls on the llm/stream waterfall, before they are billed. Auto-downgrade past a soft limit.
Two-tier image reading for text-only models in DeepSeek Harness: fast local OCR (RapidOCR, offline) first, then a vision model via modlens as fallback.
Plugin Cordis para o DeepSeek Harness v0.1: interceta a camada HTTP (auth barrier sobre o fallback da Web UI) e orquestra subprocessos de longa duracao com ciclo de vida reversivel.
Minimal DSH web GUI plugin that raises native browser notifications or fallback toasts for approval, question, plan-review, and completed-session reminders.