dsh「专家团」bundle：一句自然语言组建角色化多智能体「数字员工」团队，按九阶段门控流水线协作交付，实现者直接改代码并产出持久工件；带 live 团队浮层。 · Role-based multi-agent expert team for DeepSeek Harness: one sentence in, a staged and gated team delivery out.

:star: AI-assisted development workflow that produces understanding alongside code. Teach, capture decisions, document, assemble reviews.

Turn a one-line idea into a branch with tested, reviewed, committed code. The brainstorm-to-commit pipeline for Claude Code.

Phase-based AI development framework for AI Coding Agents. 17 specialized agents across 6 phases, handing off through files in your repo — and it reads the codebase you already have.

The only browser automation that bypasses anti-bot systems. AI writes network hooks, clones UIs pixel-perfect via simple chat.

An omni-plugin for harness engineering workflows with multi-agents, programmatic gates and skills.

HttpRunner 是一款开源的 API/UI 测试框架，简单易用，功能强大，具有丰富的插件化机制和高度的可扩展能力。

GSD Pro — The most powerful AI coding workflow for Claude Code. Free alternative to Cursor Composer, Copilot Workspace, Devin & Bolt. Multi-model routing, rollback/recovery, adaptive context. Fork of 30K★ original. npx get-shit-done-cc

An agent skill for building premium, immersive, scroll-driven websites. Works with Codex, Claude Code, and other coding agents. Also available as a Claude Code plugin.

MCP plugin that saves 98% of your context window. Works with Claude Code, Gemini CLI, VS Code Copilot, OpenCode, and Codex CLI. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY LLM: Claude/ChatGPT/Gemini on your subscription (no API key), free local models via Ollama (fully offline),

The self-hosted AI studio: local video, image, music, voice, LoRA training, coding swarms, RAG and screen agents on one GPU, driven from the Studio or by your coding agent (Claude Code, Cursor, Codex, OpenClaw) through MCP and skills.

A markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.

Active Directory pentest methodology for Claude Code: skills, agents and slash commands for internal AD red-team work (Kerberoasting, ADCS ESC1-17, DCSync, ACL abuse, NTLM relay, delegation), with per-technique OPSEC/telemetry notes. Drives netexec, impacket, certipy, bloodyAD, BloodHound CE.

Persistent visual cache for LLM-driven software development. Caches screenshots using perceptual hashing, vector search, and AX trees to prevent token overhead and visual hallucination loops.

Model-driven context management (Active Context Pruning / ACP) for the DeepSeek Harness — the model decides when and what to compress. Ported from billion-context-pi (ranxianglei); acp-kernel reused verbatim. CompactionEngine backend with compress/decompress/search_context/acp_status tools.

Windows 上让 agent 的每条 shell 命令改经 Git for Windows bash 执行;物化四个内置预设的 Git Bash 变体,统一 MSYS 盘根路径方言,侧栏终端接管可开关。· Run every agent shell command through Git for Windows bash on Windows; materialize Git Bash variants of the built-in presets, unify the MSYS drive-roo

Open-source desktop SQL workspace with 3 built-in database drivers and 16 shipped plugins, including SQL Server, DuckDB, ClickHouse and Redis. Built-in MCP server for Claude, Cursor and Devin, SQL notebooks and visual EXPLAIN.

Multi-agent mathematical problem-solving & verification frameworks for DeepSeek Harness — FOUR agent presets in one install: vibe-math-v2 (probability-driven, classic: qs.json + Propos knowledge base + explorer→solver→review/debate verdict), vibe-math-v3

DeepSeek Harness plugin for Android — build, run, and interact with a live emulator or USB device stream inside a conversation, driven entirely through adb.

Make any repo AI-first - write sustainable code from the start, or refactor a legacy codebase to prepare it for agent-driven development.Building blocks for Claude Code: subagents, slash commands, hooks, and workflow patterns. Copy what you need. A working developer's stack for Claude Code.

DSH permission-gate for the DeepSeek Harness 0.1.5 line: a single self-sufficient, deterministic-first, fail-closed gate covering P0 hard-deny -> P1 session grant -> P2 static rule (allow/deny) chain -> P3 optional LLM semantic classifier -> P4 ask, with

GitHub Issues-backed agent orchestration for Claude Code – a single Claude wears persona hats (Orchestrator, CEO, workers) routed by issue labels, driven by a heartbeat loop

`HarnessV1` adapter backed by [`@anthropic-ai/claude-agent-sdk`](https://www.npmjs.com/package/@anthropic-ai/claude-agent-sdk), which drives the `claude` CLI. The adapter ships a bridge process that runs inside a sandbox and talks to the host over a WebSo