dsh-tool-policy
Allow, ask, or deny DeepSeek Harness tool calls before execution
208 results
Allow, ask, or deny DeepSeek Harness tool calls before execution
Run the DeepSeek Harness agent's filesystem, bash and PTY terminals inside a Docker container, so execution state survives a harness restart.
DSH 容器与镜像管理插件:走官方 subprocess 服务,argv 无 shell 注入,exec 与破坏性操作默认审批。
DSH plugin: connect to a NebulaGraph 5.0 server and execute GQL queries, returning results as structured rows plus an ngql-style table render. Bundles the gql-query-generator skill.
配置打包器(Agent config packer for DeepSeek Harness):把本地 Agent 资产(Skills/会话/Profile/设置/记忆)按模块打包成 zip——迁移、分享、恢复、差异对比、隐私安全扫描,全部可选。v0.2.4:全链路异步 IO(fs/promises + execFile,不再阻塞事件循环,对外 API 一律返回 Promise)+ /packer/api/* 防护(鉴权默认 fail-closed:只用官方 connection.requestRejection,拿不到就 403;请求体上限 8MB 与速率限制 60 次/分钟在完整缓冲请求体之前判定;错误文案脱敏)、测试补齐;v0.2.3:恢复安全加固(目标路径 containment / 缺指纹 fail-closed / 备份+原子替换+失败回滚 / tar 成员白名单 / SQLite 库默认排除)+ 隐私扫描补齐(Unix 与 UNC 路径、无引号密钥与密钥形状、全量计数、无扩展名文本)+ peerDependenciesMeta 标记宿主内建 peer 为 optional;v0.2.2:管理面板按打包工作流设计语言(阶段流程条/器材面板/diff 色带/橙琥珀青品牌色)+ 深色适配(DSH 主题跟随)
Two-tier model router for DeepSeek Harness — automatic execution/judgment routing with LLM Judge, multi-model fallback chains, and exponential-backoff runtime failover. A DSH adaptation of pi-shift-router.
通过 QQ 消息远程控制 DeepSeek Harness:命令执行(/exec)、任务派发(/ask)、阶段进度汇报、屏幕截图发送(/screenshot)、AI 聊天(/chat)与聊天特化会话(/chatbind)、会话标题化管理(/newsession /title /session)。OneBot 11(NapCat/Lagrange/go-cqhttp),零 npm 依赖。
nono (Landlock/Seatbelt) backend for the DeepSeek Harness sandbox seam with a vendored fail-closed executor
Require a one-shot user approval before configured tools execute
Windows-friendly minimal agent preset for DeepSeek Harness: sandboxed pwsh, native cmd.exe, and str_replace_editor.
Content-addressed audit indexes across independent DeepSeek Harness evidence producers
Blaxel sandbox execution plugin for DeepSeek Harness (DSH)
DSH 插件: pwsh executor 增强,从 workspace .env 注入环境变量到子进程
A stateful, Agent-controllable Jupyter workspace for DeepSeek Harness: a persistent ipykernel runtime the DSH agent can read, edit, execute, and inspect, with VS Code-aligned cell UI and .ipynb save/load.
Ward Council — oh-my-pi advisor subsystem ported to DeepSeek Harness: independent reviewer models watch your agent's live transcript, investigate with read-only tools, and advise it. Advisors guide. The ward protects. The model executes.
DSH Web 插件:基于 Tauri 2 的 DeepSeek Harness 桌面启动器,自动联动桌面快捷方式,exe 路径与探测窗口可配置。
Remote-SSH-style workspace for DeepSeek Harness (DSH): pick an SSH host and a remote directory as a native DSH workspace; the agent operates the remote filesystem directly via rw_* tools (SFTP/exec). No mirror, no sync — the remote is the source of truth.
Indexed cross-session search for DeepSeek Harness — search, list, and read past and current DSH sessions through the built-in sessionQuery service. Zero runtime dependencies.
DolphinDB integration for DeepSeek Harness: the WebSocket executor, the model-facing query/write tools, and the bundled DolphinDB skills as one installable profile bundle
DeepSeek Harness (Cordis) plugin: control Alibaba Cloud ECS instances through the local Workbench CLI — list/exec/log/upload/download/diagnose/deploy/session model tools plus a visual settings panel
Task Flow / Execution Graph plugin for DeepSeek Harness (DSH): visualize a single task's full execution — turns, LLM steps, tools, skills, subagents, retries, parallel branches — as an interactive DAG with live updates and trajectory linkage.
Expose DeepSeek Harness agent capabilities as an MCP server, so an external MCP client (e.g. Hermes) can drive Harness to execute coding tasks. Hermes = brain, Harness = arms.
NeevSandbox execution providers for DeepSeek Harness
DeepSeek Harness plugin: Lord/Serf agent protocol 0.5 skills (file-based multi-agent orchestration with bounded execution and schema-validated submissions)