Plugins

Opt-in DeepSeek Harness Skill-only bundle for HOWILLMAKEIT's agent skills (undress: 技术工作去包装).

Import Cursor, Codex, and Claude Code sessions and skills into DeepSeek Harness

DeepSeek Harness bundle that registers Claw Orchestrator as an ACP subagent provider — delegate a dsh subagent to a multi-engine council across Claude Code, Codex and Cursor.

Import chat history and memory from other AI agents (Claude Code, Codex CLI, Gemini CLI, Aider, Cline, OpenCode, Hermes Agent, OpenClaw) into DeepSeek Harness

DeepSeek Harness community Bundle for local, evidence-backed Agent Memory

Claude-Code-style artifacts for dsh: an `artifact` tool that writes self-contained HTML, a bundled authoring skill, a sidebar nav button, and an in-app artifact browser with sandboxed HTML preview

Claude Code-style permission rules engine: hard/deny/ask/allow tiers with a hard tier above full access, workspace-scoped rules, wildcard path protection, and a visual staged editor; rules persist in settings.yaml.

Reuses hooks declared in Claude Code, Codex and opencode configs on dsh, executing shell, webhook, oracle and proxy handlers with zero runtime dependencies.

Chain-of-thought leakage linter: zero-dependency CLI that finds AI-session residue in docs and comments (dead design citations, PR vantage, change narration, review choreography), shipping the cot-trim fixing skill as the installable bundle.

Matt Pocock's 25 skills fully translated to Chinese (technical terms kept in English with glosses).

Independent browser acceptance testing for agent deliverables: JSON spec in, real-Chromium verdict out.

Headless Claude Code consultations with advisor, reviewer, and designer roles, returned as reference answers.

Connect a sub2api gateway to DeepSeek Harness: OpenAI-compatible multi-provider routes (OpenAI / Claude / Grok / Gemini) behind one base URL, with per-key model discovery, usage lookup, and global vision/image tools.

Animated Claude-style reasoning-effort slider and model picker that replaces the native DSH model selector.

Hooks de ciclo de vida no estilo Claude Code: comandos de shell configurados são executados antes/depois das chamadas de ferramenta do modelo, com um payload JSON via stdin; uma saída diferente de zero no pre-tool bloqueia a chamada