Plugins

Swaps the filesystem and bash providers for a mirage virtual workspace: file tools and shell commands run over mounted resources (RAM, S3, Redis, Slack, Gmail, Notion, Postgres) instead of the host disk, with per-mount read/write/exec modes, per-command sandbox routing (monty, pyodide, quickjs in process; docker, e2b, daytona remote), and installed CLIs (git, gh, slack, linear, ntn, gws, or one you register) as head words in the virtual terminal.

Open-Source-Agent-Runtime für beliebige Modelle, CMA-kompatibel, mit MCP-Tools, sandboxed Sessions, Audit, Replay und lokaler Konsole. Enthält ein natives DeepSeek-Harness-Bundle über stdio MCP.

Generative UI direkt in der Unterhaltung: Das Modell rendert interaktive HTML-Karten in den Chat-Stream, mit Streaming-Vorschau und sandboxed Rendering.

Second-model auto-review on the approval answerer chain: a read-only reviewer subagent returns structured allow/deny verdicts with reasons, fail-closed by default.

Erstellt und verwaltet sandboxed JavaScript-Tools mit einem Monaco-Editor und einem modellgesteuerten Tool-Lebenszyklus.

Rule-gated automatic approval for DeepSeek Harness sandbox escalations with an optional LLM reviewer and native human fallback.

Makes Minimal mode work on Windows by supplying the missing win32 process inspector, so the persistent shell actually starts. Adds a busybox-based preset whose persistent shell survives the workspace-write restricted token (MSYS shells die there), ConPTY console-list foreground resolution, GBK/UTF-16 file reads, and an install-trap doctor.

Login-Gateway für die DSH Web UI: Passwortschutz mit Ersteinrichtung, bcrypt plus Verschlüsselung im Ruhezustand (AES-256-GCM/HMAC), Brute-Force-Sperre, Audit-Log, TLS 1.2+ mit 80→443-Weiterleitung, CSRF-Schutz und Anti-Framing.

MCP bridge that lets ChatGPT Web create, view, continue, and supervise DeepSeek Harness agent sessions and goals while preserving DSH's native approval, sandbox, and workspace security model.

Session-aware sandbox escalation compatibility plugin for DeepSeek Harness/DSH第三方模型会话沙箱升级兼容插件

Konservatives Auto-Genehmigungs-Preset für Sandbox-Eskalationen in DeepSeek Harness

Ein deepseek-harness-Plugin für die Tensorlake-Sandbox.

Fine-grained permission gateway: per-category tool-call review (outside-workspace directories, commands, file read/write, subagents, repeated actions) with global & per-project allow/deny exceptions, quick-tool defaults, custom rules, a bilingual approval modal with inline diff details, custom rejection reasons and a sandbox-upgrade flow.

LLM auto-approval for sandbox escalation requests, with presets and a fail-closed fallback.

Git Bash shell tool for Windows: runs commands through the bash bundled with Git for Windows, with timeout, sandbox, output truncation, and background jobs, replacing the slow pwsh tool and the WSL-only bash tool.

Unterstützung für das nono-Sandbox-Backend.

Unterstützung für das microsandbox-Backend.

Lets sandboxed agents reach the web through your local proxy: net_fetch fetches pages, net_proxy_status reports proxy status.

Unterstützung für die plattformübergreifende Microsoft-Sandbox.

Self-protection guardrails for the DeepSeek Harness: protected-path interception, backup-before-destroy, CLI self-recovery commands, and strict sandbox defaulting

Alternative WorkflowEngine provider for DSH 0.1.0-rc.7 that runs model-written orchestration in fresh QuickJS/WASM runtimes with a JSON-only host bridge and bounded memory, execution time, and child-agent fan-out.

Chinese localization for tool results: intercepts tool execution output and translates English markers (`[exit code]`, `[timed out]`, `[sandbox: ...]`, etc.) into Chinese, with a settings toggle.

Press Enter to approve the sandbox permission-escalation prompt in the DeepSeek Harness WebUI - no mouse click required.按 Enter 即可批准 DeepSeek Harness WebUI 中的沙箱权限提升审批窗口，无需鼠标点击。

NeevSandbox execution providers for DeepSeek Harness