dsh-permissions
DeepSeek Harness 权限规则引擎:hard/deny/ask/allow 四级规则(hard 高于全访问)、全局与 workspace 作用域、通配符路径保护、可视化草稿式编辑器,规则持久化于 settings.yaml
2441 results
DeepSeek Harness 权限规则引擎:hard/deny/ask/allow 四级规则(hard 高于全访问)、全局与 workspace 作用域、通配符路径保护、可视化草稿式编辑器,规则持久化于 settings.yaml
Out-of-tree dsh web-profile bundle: enables --host 0.0.0.0 and gates the Web UI behind a --web_token access token (cookie after first ?web_token= open). Forks @deepseek-ai/dsh-host-webserver and @deepseek-ai/dsh-web-app/startup.
DeepSeek Harness plugin: 在侧边栏展示当前 DSH 中配置的所有模型 Provider 的账号余额(余额面板 + 底部快捷入口)
DSH composer model seat with accessible per-provider collapsible groups.
Zero-dependency MCP client bridge for DeepSeek Harness: connects stdio/HTTP MCP servers and auto-registers their tools for the agent.
L1-L2-L3 three-tier memory agent harness: gate approval, zero-judgement event log, weekly distillation, TDD execution loop. Claude Code plugin and DeepSeek Harness (dsh) bundle.
DSH model-facing delegation tool that routes continuable/one-shot subagents to a model-selected LLM provider and model
Cloudflare Tunnel plugin for DeepSeek Harness: expose the local web GUI through a quick / token / API tunnel behind a token-gated loopback proxy, with optional Cloudflare Access email OTP.
DeepSeek Harness plugin: direct multimodal image transfer declarations + per-session image clearing, reasoning-effort auto-fill, and progressive Figma MCP bridging (design-to-code + AI-driven design).
DSH client plugin: 检测回合被中断/出错/超长时, 自动发送「继续」重试, 不做模型/provider 切换。带设置卡片。
编程陪伴 Node-host companion: opens a persistent, per-site popup window while agents work (minimizing + suspending its process tree to freeze playback while they wait), with multi-site profiles and an in-window return-to-chat button.
YiziMarkdown theme pack for DeepSeek Harness — 19 style themes with custom brand and prompt mapping.
DSH 救援恢复插件(2.5.0:profile 变化提交 git 时自动打包 zip 还原点——原始相对路径手动 unzip 覆盖恢复,文件名后缀标注触发插件):守护进程 + .dsh git 仓库管理 + 远端备份(token/SSH)+ 救援环境 + 专项工具 + git 还原(只还原 profile)+ 插件安装门禁 + 自动更新(大版本/结构不一致卸载重装)
A personal DSH browser plugin: persistent profile, visible window, human-in-the-loop login, and safe agent browser tools.
Global model request headers plus image input, reasoning, and DeepSeek system-role compatibility for custom providers
In-process DeepSeek Harness plugin: chat with your DSH agent directly in WeChat
企业 AI 资源统一管理平台 —— 基于 DeepSeek Harness「一切皆插件」架构
Self-evolving plugin for DeepSeek Harness: observes agent behavior, proposes improvements to its own skills/policies via LLM, validates them in a sandboxed trial agent, and applies versioned mutations with rollback.
Busy-time input queue for the DeepSeek Harness web GUI: three-tier (now/next/later) planning, queue editing with drag-to-reorder and concurrency protection, interject (steer) and interrupt, batch clear with confirm, session freeze - a cordis client plugin, no dsh source changes, no PR required
Auditable, workspace-local project wiki for DeepSeek Harness with a native Web view
A compact per-model reasoning effort editor for DeepSeek Harness custom providers.
Subagent delegation with per-call model/provider override (host) + multi-panel live streaming subagent viewer (client)
Generic LLM gateway model provider plugin for DeepSeek Harness: mounts newapi / LiteLLM / Higress / any OpenAI-compatible gateway as provider routes, auto-discovers the model list, enriches every model with models.dev parameters, and dispatches each model over its native wire protocol (OpenAI / Anthropic / Gemini) via the pi-ai SDK
DeepSeek Harness plugin that locks the web chat composer Enter-send. Toggle with a lock button or Ctrl+Shift+K to prevent accidental sends while typing or editing long prompts; unlock to send. IME-safe and per-session.