开源工作流自动化，安装并运行

Conductor 是一个事件驱动的智能体工作流引擎，为应用程序和 AI Agents 提供持久且高韧性的执行引擎。

测试发现：Conductor server v3.30.2-SNAPSHOT builds and starts with SQLite persistence on port 8080. Workflow definitions can be registered via the REST API and workflow instances can be started, tracked, and inspected. The server responds to health checks, actuator endpoints, and all core REST API endpoints. 17 modules pass... 19 分钟。

Cadence 是一个分布式、可扩展、持久、高可用的编排引擎，用于以可扩展且弹性的方式执行异步长时间运行的业务逻辑。

测试发现：Cadence server compiled all 8 binaries, installed SQLite persistence schemas, started all 4 services (frontend, matching, history, worker) into RUNNING state, and passed all unit tests across the codebase. 23 分钟。

FlowGram 是一个可扩展的工作流开发框架，内置画布、表单、变量和素材，帮助开发者更快、更简单地构建 AI 工作流平台。

测试发现：FlowGram monorepo builds successfully (73/73 Rush projects), all tests pass across 9 packages with vitest (0 failures), and the demo-free-layout app serves correctly via Rsbuild dev server answering HTTP 200. 15 分钟。

将长期目标转化为可验证的结果。一个面向长期运行的 agent 团队的本地优先语义控制平面和研究工作台，跨会话和运行时。

测试发现：LoopX 1.2.0 installed from source in venv with Node.js 22; CLI responds, doctor passes, demo workspace serves on localhost, 125/125 TypeScript control plane tests pass, canary and smoke suites pass, main Python test suite collects 12206 tests with 3 pre-existing failures (stale registry manifest, missing git worktree... 59 分钟。

开源、自托管的 AI agent，端到端运行 YouTube 频道：研究主题、撰写脚本、生成旁白、视觉和视频、优化 SEO，并在您批准后发布，然后从分析中学习。支持 Gemini、OpenAI、OpenRouter、DeepSeek 等。

测试发现：npm install completed, all 49 system tests pass, and the dashboard server starts and responds 200 on /health. 8 分钟。

用于使用 LLM 为 AI IDEs 生成 PRD、技术设计和 MVP 等的模板与工作流程

测试发现：The vibeworkflow CLI builds and passes all 27 unit tests; the scaffold correctly fills app name, one-liner, target users, phase, tech stack, and commands from PRD/TechDesign metadata; the CLI detects tools, generates AGENTS.md, CLAUDE.md, agent_docs, and skills, and surfaces remaining placeholders. 6 分钟。

YAO = Yielding AI Outcomes。一个用于可复用 agent 技能的严格工程、评估、治理和可移植性系统。

测试发现：All 87 CI tests pass, the CLI validates the project correctly, and the Review Studio reports decision 'review' with score 86. 21 分钟。

用于 Kubernetes 的事件驱动自动化框架

测试发现：Argo Events CLI binary builds and runs, all 78 unit test suites pass with 0 failures. 12 分钟。

一个运行时基础层，将 agent 的执行转换为可逆的、类似 Git 的追踪记录，以便 meta-agent 可以观察、fork、重放和回滚任何运行。将 agent 和环境结合在写时复制 fork 中，比 docker commit 快约 5x，并且在重放时约有 95% 的 KV-cache 复用。为...

测试发现：All 50 workspace packages installed via uv sync; integration test suite reports 77/82 passed (4 skipped due to missing native jail, 1 expected failure for fuse-overlayfs check); core package tests 553/554 passed (1 slow-hypothesis flake); shepherd2 tests 186/186 passed; offline quickstart runs successfully end-to-end. 33 分钟。

CLI 任务管理与自动化工具

测试发现：doit v0.38.dev0 installed and fully functional, all 830 tests pass and the CLI executes incremental build tasks correctly with file dependency tracking, up-to-date skipping, and clean operations. 2 分钟。

自托管的 Web 应用，可自动化下载、整理和调度 YouTube 频道内容，支持 Plex、Kodi、Emby 和 Jellyfin

为你的 AI agent 赋予 n8n 超能力。537 个节点（含完整 schema）、7,700 多个模板、类 Git 同步以及 TypeScript 工作流。

测试发现：npm install --ignore-scripts completes with warnings only, tsc -b TypeScript compilation succeeds, CLI version 2.7.0 responds to all commands, and 6 test suites pass (922 total tests, 0 failures across transformer, CLI, skills, MCP, openclaw, and vscode packages). 19 分钟。

RuleGo 是一个轻量级、高性能、嵌入式、下一代组件编排规则引擎框架，适用于 Go。

测试发现：The rulego Go library compiles with go build and all its test suite passes (go test exits 0, PASS). 8 分钟。

一站式工具，用于管理您的域名组合。有域名？就用 Domain Locker。

测试发现：Dependencies installed, all 37 tests pass, production build completes, SSR server serves /api/health (200), / (200), /login (200) correctly. 9 分钟。

开源、隐私优先的客户之声分析，是一个自托管的 Enterpret 和 Unwrap 替代方案，在任何 LLM 看到反馈之前剥离其中的 PII。

测试发现：Obsei library installs, all 45 tests pass (including real model inference with transformers), and example scripts run successfully producing analysis output. 12 分钟。

µTask 是一个自动化引擎，用于建模并执行以 yaml 声明的业务流程。

测试发现：uTask 1.24.0 builds into a working ELF binary (44MB), starts an HTTP server on port 8081 that serves the OpenAPI spec, responds 200 pong on health endpoint, and all 14 test packages pass on a fresh PostgreSQL 16 database. 16 分钟。

适用于人类和 AI agent 的开源协作系统。共享上下文、协调任务，与你的团队、云端 agent、Codex、Claude Code、Grok、Grok Bot、OpenCode、Antigravity 和 Cursor 并行工作。

测试发现：The OpenMeld monorepo builds cleanly (31/31 tasks). The openmeld CLI binary (v0.3.262) runs and prints version and help information. The public documentation dev server serves HTTP 200 on /docs. The desktop Electron application launches and its main process remains alive on the virtual display. The CLI test suite... 49 分钟。

根据你在 Jellyfin、Plex 或 Emby 上最近观看的内容，轻松向 Jellyseer/Overseer 请求推荐的电影、电视剧和动漫，让 SuggestArr 全自动处理，让你的媒体库保持新鲜，带来新颖刺激的内容。

测试发现：SuggestArr backend runs on Python 3.12, serves API on port 5000, initializes all database tables on first boot, and all 1050 pytest tests pass without errors. 3 分钟。

安全的 MCP 运行时，让 AI agent 操作本地机器、服务器和容器，支持多设备编排。

测试发现：AgentDock binary was built successfully, all 53 test packages passed (0 failures), the server started on 127.0.0.1:8765, and a JSON-RPC ping to the MCP endpoint returned HTTP 200 with a valid result. 2 分钟。

开放核心工作流引擎，驱动 Bubble Lab，并且完全可独立运行、托管和扩展。

测试发现：All 44 test files across 4 core packages and the API server pass; the API server starts on port 3001 with SQLite database migrations complete and dev user seeded. 27 分钟。

Temporal 的 Go SDK

测试发现：The Temporal Go SDK repository installs and builds with Go 1.26.0, passes its static checks, all 2136 unit tests, and the full 636-test integration suite (including real workflow/activity execution) against the embedded Temporal dev server. 29 分钟。

AI 跨境电商运营智能体工具箱 · 把运营 SOP 拆成 7 个可复用的 AI Agent 工作流，含 Prompt、模板与真实案例

测试发现：8 Python automation scripts (market research, pricing, title linting, spec consistency, ad metrics, image check) run on Python 3.12.3 with only Pillow as additional dependency, processing sample data end-to-end with correct structured output. 5 分钟。

一个用于递归 agent 的复合 agent 操作系统。同时也是 Grok Bot 和 Meta 的 Muse 的开源替代品。

测试发现：One non-root install with vendored Node v24.21.0 and pnpm 11.6.0: install, typecheck, build, and the full test:unit suite passed (two consecutive exit-0 runs), and the core server boots and serves real /api health/apps/agents/routines/action-executions responses on 127.0.0.1:4141. 54 分钟。

一个基于 Next.js 16、Baileys 和 Prisma 构建的自托管 WhatsApp 网关与仪表板。支持管理多会话账号、定时消息、自动回复，并通过 Webhooks 集成。

测试发现：WA-AKG 1.6.4 builds and runs on Node 22 against a locally stood-up MariaDB 10.11.8 (19 Prisma tables), a SUPERADMIN plus OWNER user and a session created via the real API, with the web UI, docs, and authenticated API endpoints all returning HTTP 200. 12 分钟。

面向企业的数字员工构建与运行平台：把专业员工的经验、流程与判断标准，固化为可随时上岗、可配置、可审批、可观测的 AI 数字员工。

测试发现：Backend server on port 8888 serves /health returning 200 with all 7 SQLite databases ok; 553 tests pass covering auth, catalog, conversations, approvals, analytics, security, guardrails, assignments, automations, ontology, traces, sandbox management, summarization, knowledge retrieval, tool calls, streaming, and... 38 分钟。

司南：个性化 AI 任务总控 Skills 系统 / COMPASS: Personal Alignment Skills OS for AI Agents

测试发现：All 85 tests pass, 34 Python scripts compile cleanly, CLI tools produce correct output, smoke tests pass for session-handoff-prompt and user-profile-keeper, task-forest generates valid HTML exports, and the assess-interview-candidate case-creation pipeline works end-to-end. 6 分钟。

可从小型简单项目扩展到大型复杂代码库的 AI agent 工作流工具。

测试发现：cg binary (2.5 MB, v1.3.0) compiles from source, indexes 282 files into a SQLite code graph, runs all 356 unit tests and all 40 integration tests passing, and responds to commands like --version, init, search, and serve over stdio. 8 分钟。

规范驱动开发与智能压缩，结合 Ralph Wiggum 循环和结构化规范工作流的 Claude Code 插件。

测试发现：Smart Ralph plugin suite fully passing: 342 Bats tests and 13 Python tests all green, with stop-watcher, state management, integration, prototype phases, interview framework, phase gates, workflow guardrails, and platform validation all verified. 23 分钟。

Temporal Rust 与 Core SDK

测试发现：cargo build, cargo test (982 unit tests, 0 failed), cargo lint, and cargo test-lint all pass cleanly on Rust 1.94.1 with protoc 29.3 at /home/runner/.cargo/bin/protoc. 76 分钟。

多模型 DAG 驱动的并行 AI 电影生成：并行加速随场景独立性扩展；同时生成电影场景而非逐个生成；"把影视生成的执行图从拓扑序变成关键路径最优调度"; 唯一把场景叙事依赖建模为 DAG、以 CPM 算法驱动并行调度的影视生成引擎

测试发现：Forge installs cleanly in a venv, all 20 unit tests pass, forge plan and forge run complete end-to-end with mock backends and produce a valid MP4 output file. 5 分钟。

Vibe Check 是一个工具，为 AI Agents 提供类似导师的反馈，防止在复杂和长期 agent 工作流中出现视野狭窄、过度工程化和推理锁定。告别你那些过于急切的 AI Agents。适用于：编码、模糊任务、高风险任务。

测试发现：With Node 22.23.3 at /tmp/node22, npm ci and npm run build succeed, all 139 unit tests pass, the MCP server answers 200 on /healthz and handles repeated initialize/tools/list/tools/call requests over HTTP with real LLM responses, and vibe_learn entries persist in ~/.vibe-check/vibe-log.json. 16 分钟。

Openclaw 多智能体协同系统 | 面向决策者的多智能体操作系统，基于 OpenClaw (Clawbot) + Slack，让 AI 团队各司其职、自主稳定迭代。

测试发现：OpenClaw 2026.10.5 gateway runs with 7 OpenCrew agents (main, cos, cto, builder, ko, ops, research) and real OpenRouter model inference works. 13 分钟。

自托管 AI agent 消息网关：支持 Codex、QQ、微信/企业微信、飞书、智能设备及桌面自动化。| AI 消息网关：连接聊天、智能设备与 Agent，支持定时任务和角色上下文。

测试发现：A clean-state Linux Host stayed healthy with Manager HTTP 200, businessReady=true, two seeded roles, one ready route, 29 catalog plugins, 33/33 advertised assets returning 200, and a successfully opened NapCat WebSocket, while optional speech capture and NapCat HTTP profile access remained unavailable. 45 分钟。

开源电商图片生成器、翻译与 agent 工作台。

测试发现：npm install and Next.js build succeed, dev server returns HTTP 200, and all 168 tests pass with zero failures. 28 分钟。

面向 AI agent 的本地优先游戏设计操作系统：将会话转化为证据、实验、可审查的决策和持久的项目记忆，支持 Human Gates 与回滚。

测试发现：gamedesignos 1.3.0.dev0 is installed in a virtualenv, its 89-test unittest suite passes, and a complete decision-to-accept workflow (evidence, experiment, review, Human Gate accept, VOI, gates, graph, pack) runs successfully on a created workspace. 5 分钟。

Temporal 的 Java SDK

测试发现：Temporal Java SDK compiles, passes spotlessCheck, and all non-cloud tests pass across temporal-sdk, temporal-test-server, temporal-testing, temporal-kotlin, temporal-spring-boot-autoconfigure, and temporal-serviceclient subprojects. 71 分钟。

工作流编排框架，开源，适用于您的产品：一个白标 React 工作流构建器，在您自己的 Temporal 上实现持久执行，为 AI agent 提供人工审批步骤，以及每次运行的实时视图。

测试发现：Install succeeded: pnpm install and pnpm preflight pass, all 830 tests pass across all packages, the demo builds and serves HTTP 200 at localhost:4200/6007 with the Workflow Builder editor UI, and the backend/worker/ai-studio packages compile and pass their tests. 22 分钟。

自托管 PaaS 解决方案

测试发现：Go 1.27.2 toolchain installed, all 170+ dependency packages resolved, hivepaas binary built successfully, full test suite passes (exit 0) with no failures. 12 分钟。

开源自助式工作流执行平台

测试发现：The flowctl binary builds and runs, PostgreSQL 16 serves the database, migrations apply, the server starts on port 7000, login with admin credentials succeeds, and authenticated API endpoints return valid data. 11 分钟。

agent 技能的任务控制

测试发现：LazySkills binary builds from source, passes its full test suite (including race detector), passes all CI checks (gofmt, vet, deadcode), and its non-TUI commands (doctor, find, update, version) execute correctly against both the filesystem and the external skills.sh registry API. 3 分钟。

使用 Kubernetes（Talos）、ArgoCD 和 Terraform 进行基础设施即代码的 GitOps 管理的家庭实验室。

测试发现：Portfolio: Next.js 16 standalone build produces a server responding 200 on /api/health and /; Blog: Hugo 0.166.0 builds 85 pages to static HTML. 19 分钟。

开源、本地优先的 AI agent，用于编码和实际工作，支持 BYOK 模型、MCP、技能、插件、工作流和私有知识库。

测试发现：Open Vetta monorepo installs, builds (CLI, all library packages, 14 presets), lints, passes all architecture guards, and runs the vetta CLI binary. Test suites across 39 packages pass with 2870+ passing tests covering AI providers, agent loop, runtime-core, coding-agent, MCP, storage, tools, subagents, agent-team... 15 分钟。

为 Hermes Agent 精选的真实世界用例，该自我改进的 AI agent 来自 Nous Research，并有第一手资料佐证。

测试发现：verify-usecase.py runs without errors on all 68 usecase files, generating structured verification reports with confidence scores; 6 demo shell scripts parse cleanly; CI workflows reference valid file paths. 12 分钟。

一个能记住自己做过什么的 cron：一个 CLI 作业调度器，具有持久运行历史、stdout/stderr 捕获和实时终端仪表盘。

测试发现：The rote binary builds from source with Go 1.27.3, all 8 test suites (522 cases) pass, the binary runs jobs on demand, records results (exit code, duration, stdout, stderr) to a SQLite database, and lists/logs history correctly. 5 分钟。

类型安全、事件驱动的工作流编排库，可随业务扩展。使用 Go 构建健壮的分布式工作流，具备编译期安全、自动重试和开箱即用的水平扩展。使用适配器与现有技术栈集成。

测试发现：The repository builds from source with Go 1.26.10, all internal unit tests pass (14 packages, 0 failures), and three in-memory example workflows (gettingstarted, callback, timeout, schedule, connector) execute end-to-end successfully. The embedded Web UI server starts, serves the homepage, and responds to all 3 API... 4 分钟。

行业与公司研究领域包，适用于 DeepSeek Harness：方法论技能、产业链映射、公开来源政策/新闻追踪、公司研究卡片以及可审计的研究报告。仅用于研究，不构成投资建议。

测试发现：Project installs, builds, typechecks, passes all 141 tests, and passes the full 12-command verification gate (typecheck + typecheck:ci + typecheck:checkout + test + build + verify:self-contained + verify:artifacts + verify:readme-sync + verify:skills + check:lockfile + pack:check) on Node 22.23.4 with pnpm 11.7.0. 2 分钟。

用于 DeepSeek Harness 的可验证研究报告引擎：内容寻址的证据账本（声明-快照绑定，防篡改）加上带版本控制的密封报告，包含每个声明的验证结论以及清单密封的目录。

测试发现：dsh-research-report v0.3.21 is fully built, typechecked, and tested: 111/111 tests passing across 16 test files, all CI verification scripts green, and the CLI binary responds to --help. 2 分钟。

Compartment 是一个自托管部署系统，用于在团队控制的基础设施上部署小型软件。

测试发现：The monorepo installs, builds, and all test suites pass (2832 tests across scripts, CLI, SDK, contracts, edge, worker, API, and console). The CLI binary responds to --help. The full app stack (pnpm dev) requires PostgreSQL, caddy, buildctl, docker, and railpack which are not present in this container. 18 分钟。

驱动 Flowbaker 的执行引擎，一个用于构建、连接和自动化工作流的无代码平台。

测试发现：Flowbaker Executor builds from source with Go 1.27.3, its test suite passes entirely (3 packages, 142 assertions), and the binary responds to all CLI commands (help, status, start). 4 分钟。

开源、自托管的 OpenAI Dots 和 Grok Bot 替代品。专为不被反机器人系统检测而设计。

测试发现：npm ci installed 428 packages, the CLI bundle and web client built, npm run typecheck passed, npx vitest run gave 2103 passed/3 skipped and go test gave 120 passed after I fixed the secret-input backspace bug and the execEnv DISPLAY bug, and the control plane served HTTP 200 on /api/health, /api/dots and the web UI... 22 分钟。

可定制的无代码组件，用于构建基于流程的编程应用或工作流自动化。0 个外部依赖项。

测试发现：Core designer builds completely (lib/cjs, lib/esm, dist/umd) and passes all 112 unit tests via karma+jasmine on Chrome Headless 130. 13 分钟。

Second Brain 是一个 agentic 框架，充当操作系统，利用本地文件智能、工作流自动化和 LLM 来完成任务，并通过多种模态和消息平台进行通信。

测试发现：Second Brain Python kernel installed and starts. HTTP frontend serves port 8789 and responds to authenticated SDK requests. Web UI builds with Node 22. Test suite runs 2819 passing tests with 2 pre-existing failures. Requires Node 20.19+ or 22.12+ for the web UI; port 8787 is blocked by an orphaned socket from prior... 51 分钟。

用于定义和运行持久化工作流与活动的核心软件包。支持长时间运行的持久化工作流、重试、队列、并行执行、工作流监控、专用存储连接，以及面向微服务、数据管道、saga、agentic 工作流和其他复杂业务流程的编排。

测试发现：composer install completes with no errors; composer unit runs 2098 tests with 15727 assertions, 3 cosmetic failures in ProjectionPrefetchTest due to PostgreSQL 16 trailing-zero timestamp format, no errors, 3 skipped, 15 PHPUnit notices. Static analysis PHPStan requires --memory-limit=-1 due to container memory... 43 分钟。

可破解且自托管的开源替代品，用于 Vercel、Heroku、Netlify 等。所有应用程序集中在一处。使用您喜爱的工具进行开发和部署，并通过 GitHub 享受持续部署。

测试发现：Dependencies install and typecheck cleanly, the web app builds and its 155 tests pass, the API passes 322 of 326 tests (4 pre-existing request-validation assertions fail), and a running API served HTTP 200 on /server/health from its own Postgres 16 instance while the built web app served HTTP 200 from vite preview. 6 分钟。

构建你可见、可编码、可运行且可拥有的工作流。开源可视化工作流自动化，使用真正的 TypeScript、隔离执行和自托管 Control API。

测试发现：Installed Bun 1.4.2 and Node 26.9.0; bun install succeeded with 556 packages; all workspaces built and tested with 3097/3098 tests passing (1 consistently flaky socket endpoint test in mcp.test.ts); the server started on port 3000, answered HTTP 200 on GET /, returned valid Control API and MCP responses, and created... 14 分钟。

面向生产环境 AI 的多 agent harness

测试发现：With the fixes, ./quickstart.sh completes and launches the Hive dashboard on an available local port; /api/health returned 200, Chromium rendered the Hive landing UI without page errors, and make test exited 0 with 2496 core and 4381 tools tests passed. 21 分钟。

智能技能路由与工作流编排，面向 AI agent：在 SkillsBench 上使用 DeepSeekV4Flash-VE 时，+21.12 pp 奖励，tokens −29.6%。

测试发现：VibeSkills v4.1.0 installs successfully, check verifies all 292 receipt-owned files, the vgo_cli Python launcher works end-to-end, and 129+ core unit tests pass with 3 isolated test failures due to missing pwsh in this container. 37 分钟。

Temporal 服务

测试发现：Temporal server v1.33.0 built with Go 1.27.3 from source; starts on SQLite in-memory exposing all gRPC/HTTP endpoints; 241/244 unit tests pass (3 pre-existing flaky matching-engine backlog timing tests fail). 11 分钟。

本地优先的开源 AI 助手，用于管理你的数据。统一管理任务、笔记、文档、照片和书签。私密、自托管，并通过 API 可扩展。

测试发现：On this container, Eclaire installs and builds (pnpm install, app:upgrade to 0.7.0 on SQLite, pnpm build all packages, typecheck 16/16), 2185 unit tests pass across all six packages, the backend serves HTTP 200 on /health and the SPA, the CLI status/settings/user commands work against the live backend, and the... 58 分钟。

Grasp 帮助你在一个可视化工作流中管理多个项目和并行的 coding agents，从而让人更快理解并交付更多。

测试发现：With user-local Go 1.27.3 and Node 24.12.0, the built Grasp server and gateway binaries start and answer health 200, admin login and project APIs return 200 with seeded data, and the server, web, docs, gateway, and sandbox build, test, and coverage suites all pass. 18 分钟。

由 AI 驱动的 StartUp Accelerator Engine，使用 Next.js、LangChain、PostgreSQL 和 pgvector 构建。上传、整理并与文档聊天。包括预测性缺失文档检测、基于角色的工作流和页面级洞察提取。

测试发现：LaunchStack monorepo installs cleanly after Node upgrade to v20.11.0; package-level vitest (327/327) and web jest (hundreds of unit+integration tests) all pass against a standalone PostgreSQL 16 with pgvector; only missing real external credentials for chat, S3, and provider APIs. 33 分钟。

开放工作流智能，用于跨框架发现、验证、修复和生成自动化工作流，从 n8n 开始。

测试发现：All 21 Python tools in tools/ run with standard-library Python 3.12 with no dependencies or credentials: audit scans 3181 workflows, search returns ranked results, prompt-to-n8n exports valid workflows, static validation passes, repair/learning pipelines work, and schema validation confirms IR/catalog/metadata... 3 分钟。

使用 OpenClaw + Google Gemini + Docker 免费搭建 AI Telegram Bot

测试发现：OpenClaw Setup v5.17.5 Web UI responds with HTTP 200 on root, serves all static assets, and returns correct system metadata from /api/system endpoint, running on a standalone Node 24.0.0 binary that provides the required node:sqlite built-in module. 6 分钟。

Agent-ready RPA 套件，提供开箱即用的自动化工具。为个人和企业打造。

测试发现：Engine Python 3.13.15 dependencies installed, engine modules import correctly, 207 pytest tests passing across data processing and encryption components, 1 source bug fixed (string fill slice type). Frontend shared and CLI packages build. Can't run backend services or Electron desktop app without Windows/Docker. 22 分钟。

在你自己的 PC 上 24/7 工作的自主公司：Windows/Linux/macOS。

测试发现：The dashboard and product-center HTTP servers, the Python test suite (619 of 632 passing; 13 isolation-blocked failures), all fake-engine shell integration tests, frontend JS contract suites and example Node tests except scopefence run successfully in this container. 18 分钟。

在一台全新的 Ubuntu 或 Debian VPS 上，通过一条命令安装你自己的 AmneziaWG 3.1 VPN 服务器。专为封锁 WireGuard 的网络设计。无需 Docker，支持 QR 码导入，2.0 版本可应要求用于路由器。

测试发现：All 156 Bats test suites pass on Ubuntu 24.04. The AmneziaWG installer shell scripts (awg_common.sh, manage scripts, install scripts) are verified through isolated unit tests covering CIDR helpers, config validation, PPA key embedding, QR code generation (structural), all v5.x bugfixes, protocol markers, routing... 36 分钟。

适用于 EHRs、支付方门户以及你的团队点击使用的任何应用。一个人或一个 agent 演示任务；OpenAdapt 将其转化为本地程序，无需模型调用即可重复步骤，检查每次保存，并在不匹配时停止。

测试发现：Package installs and all 288 tests pass; CLI, doctor, version commands work; recording generates valid artifacts; compile is CPU-bound by onnxruntime in this container but eventually produces bundles. 20 分钟。

三省六部制 · OpenClaw Multi-Agent Orchestration System：9 个专业化的 AI agent，具备实时仪表板、模型配置和完整的审计轨迹

测试发现：All 68 tests pass on Python 3.12.3; the dashboard HTTP server serves the React frontend, health check, and all API endpoints correctly. 6 分钟。

一个在您自己的 PC（Windows/Linux/macOS）上 24/7 运行的自主公司。

测试发现：All 419 Python tests pass, all 4 shell test suites pass, all 19 frontend JS tests pass, dashboard server starts and serves HTTP 200 with structured JSON on / and /api/status. 9 分钟。

让你的命令行 AI agent 在后台自己接活、自己跟进到底；平时不烦你，遇到拿不准的事才发卡片找你拍板。

测试发现：The project installs and runs: the hub answered 200 on /healthz, the real Codex CLI completed a full ask-answer loop in the same resumed session reporting '试跑通过', test:quick passed 280/282, typecheck and format:check passed, and install-smoke.sh verified install-to-uninstall. 56 分钟。

clodfarm（大声读出来）：一个 Claude Code agent 农场。种下一个任务，它们会将其拆分为 sub-agents，打开工作，并根据每个账户真实的 5 小时和每周使用量来调节节奏。从 Claude app 中操控它。

测试发现：clodfarm 1.13.0 installed and its full test suite passes (411/522 tests, 111 skipped due to missing Claude Code binary); CLI responds with version and help output; SQLite and DynamoDB store backends are functional. 30 分钟。

开源免费的自托管 SSL 证书 ACME 工具，可视化地自动化完成签发、部署、续期和监控的全流程。

测试发现：Certimate v0.4.33 builds from source with Go 1.26.8, serves web UI responding HTTP 200, and runs all 3 test packages successfully. 17 分钟。

一个简单的 CLI 工具，用于自托管 Next.js 应用程序。

测试发现：Quicky v0.3.0 CLI responds to commands, all 13 npm dependencies installed, PM2 7.1.0 daemonized and managing processes, config/release/validate/health modules verified, and the app is ready for use with GitHub credentials and system packages (nginx, certbot) configured on a server with root access. 7 分钟。

你的轻量级开源个人 agent。

测试发现：After pinning mcp<1.30 and relaxing two width-sensitive CLI assertions, Python installs editable, the 6785-test suite, 1125 WebUI tests and basedpyright all pass, and against a local mock provider the CLI agent, OpenAI-compatible API server (JSON and SSE), gateway WebSocket chat and native TUI all returned 'Hello from... 29 分钟。

ODS V3 预发布版：在正式 V3 发布前进行公开测试和优化。将你的 PC、Mac 或 Linux 设备变成私有 AI 服务器。

测试发现：Project builds correctly, all 418 bats tests, 14/14 smoke tests (after ai_err fix), ~300 shell/Python contract tests, and the vite frontend build pass. The Docker runtime install cannot complete in this environment, but the dry-run installer, CLI (-v2.6.0), all standalone test suites, and the frontend are functional. 15 分钟。

AI agent 技能用于社交媒体研究。涵盖 TikTok、Instagram、YouTube、Reddit、X、LinkedIn 等平台上的异常帖子、评论挖掘、竞品拆解、广告库和趋势。由 ScrapeCreators 提供支持。适用于 Claude Code、Cursor、Codex、Gemini CLI。

测试发现：All 13 SKILL.md files have valid frontmatter and non-empty bodies, the npx skills CLI installs the repo correctly (13 skills listed under .agents/skills/), and the validate-skills.py script passes cleanly. 3 分钟。

Claude Cowork 的开源版本，支持 500+ 个 SaaS 应用集成

测试发现：Backend server starts and responds on all API endpoints (health 200, chat 200 with SSE streaming, abort 200, validation errors 400); Electron desktop app launches and prints 'Electron app ready'. 10 分钟。

受中国治理模式启发的开源 multi-agent 协作系统，使用 OpenClaw 部署和协调专门的 AI agent。

测试发现：npm dependencies install cleanly, all 19 Jest tests pass, both manual test suites pass, OpenClaw Gateway runs and responds HTTP 200 on dashboard and health endpoints with a mock LLM provider configured. 19 分钟。

自然语言驱动的无限画布工作流 Agent，让 AI video 创作第一次真正变成可编辑的工作流。 AICON 面向创作者，提供从剧本拆解、分镜生成、素材生成、video 合成到内容分发的一整套能力。 不是只给你一个输入框，而是让你用自然语言和无限画布一起驱动创作，把文本、图片、video...

测试发现：Backend FastAPI server starts and serves root, health, and API v1 endpoints on port 8000. Frontend builds to dist/. 57 backend unit tests and 15 frontend utility tests pass. 22 分钟。

Autoprompt 是一种编码 agent 技能，可将 agentic 编码任务的失败率降低 45%。

测试发现：autoprompt-skill v2.0.0 CLI installs globally, all 256 core/provider/benchmark tests pass cleanly on Node.js 20.20.2 with Python 3.12 + PyYAML 6.0.3 + jsonschema 4.26.0. 31 分钟。

通过 Slack 最大化 Claude Code Pro 使用的 24/7 AI agent。自动处理任务、管理隔离工作区、创建 Git 提交/拉取请求，并优化白天/夜间使用阈值。

测试发现：Installed and built cleanly in venv; all CLI commands (think/check/report/cancel/trash) work; sle daemon runs the full Slack-mode loop (socket-mode broker, /check success 200, mock task execution COMPLETE) against local mocks; real Anthropic/Slack credentials still needed for genuine end-to-end operation. 53 分钟。

免费、开源的可替代 Weavy AI、Krea Nodes、Freepik Spaces 和 FloraFauna AI 的基于节点的 AI 工作流构建器，用于生成式图像和视频流水线。

测试发现：Frontend Next.js app serves on port 3000, backend FastAPI serves on port 8000 proxying to a mock MuAPI on port 9000; root page and workflow listing page both return HTTP 200, backend health check returns 200, and all workflow API endpoints respond correctly through the mock. 16 分钟。

通过 Model Context Protocol (MCP) 将 AI 助手集成到 n8n 工作流自动化中。连接 Claude Desktop、ChatGPT 及其他 AI 助手到 n8n，实现自然语言工作流管理。

测试发现：Project installs and builds cleanly. All 78 tests pass (7 test suites, 0 failures). The MCP server binary starts on stdio and announces 34 tools. No real n8n instance was needed, tests use mocked axios. 5 分钟。

面向企业环境限制下的个人 AI 助手，基于您已有的 coding agent、工具、会话和权限构建。

测试发现：Python 3.12 venv installed the tenx-productivity editable package, the 16-test suite passes, and the slack-polling trigger host and stand-up prep CLI run end to end against a local schema-correct Slack mock, so the same binaries are ready for a real Slack token/cookie and a real coding-agent CLI once those credentials... 13 分钟。

用于数据工程工作流的 AI agent 工具

测试发现：astro-airflow-mcp installs cleanly (uv sync --all-extras), its unit suite passes 610/610, its integration suite passes 44/45 (1 skipped) against a schema-correct Airflow 3 mock, the MCP server answered HTTP 200 to MCP initialize and get_airflow_version returned 3.4.0, and the af CLI returned version... 49 分钟。

Metis 是一个 coding agent，可将 AI/LLM 编码性能提升 50% QQ群: 801193315

测试发现：npm ci installs cleanly, npm run build produces working dist/, CLI prints version 1.4.0 and full help, 268 of 275 test files pass in non-root headless Electron environment. 34 分钟。

一个允许你挂机收集定时 Twitch 掉宝的应用，支持自动领取掉宝和切换频道。Web GUI，Docker 化。

测试发现：The project installs with uv sync, all 692 Python tests pass, the FastAPI web server starts on port 8080 and responds 200 on /, /healthz, and /api/version, and shutdown is graceful with exit code 0. 3 分钟。

运行你自己的 agents 组织

测试发现：Huddol Rust CLI, Python core app (stdio and websocket), and webui all install and run. The core app accepts and responds to JSON-RPC requests over stdio, and listens on a loopback websocket port with token auth. The webui builds to production HTML/CSS/JS. Python test suite shows 954+ passed across 22 test files with 1... 30 分钟。

轻量级自托管 AI agent。开源 OpenClaw 替代方案，使用 Go 编写。

测试发现：GoGogot binary builds, all 4 Go module tests pass, and the agent initializes and polls for updates against a mock Telegram server with a mock LLM provider configured. 9 分钟。

通过自然语言实现 AI 驱动的 n8n 工作流自动化。MCP 服务器使 Claude AI 和 Cursor IDE 能够通过 Model Context Protocol 创建、管理和监控工作流。支持多实例，提供 17 个工具，文档全面。以对话方式构建工作流，无需手动编辑 JSON。

测试发现：The n8n-workflow-builder MCP server builds with TypeScript, passes all 21 Jest unit tests, and launches listening on port 3456 with a working health endpoint. 2 分钟。

生产就绪的 ComfyUI 自定义节点，支持 1,400+ 个 fal.ai 模型，以及自动更新的图像、视频、音频、LLM 和 VLM API，并具备原生媒体、缓存和成本控制。

测试发现：All 183 tests pass, the Python package imports and registers 1691 fal nodes correctly from the committed registry, and every static node module (image, video, LLM, VLM, trainer, upscaler, platform, billing, inbox, builder, 5 utility modules) and the dynamic node factory load without errors. 3 分钟。

可自部署的 Grok 注册机 Web 控制台，支持 Docker 与 DrissionPage 自动化，并包含邮件后端配置和本地账号号池管理。

测试发现：The web console builds, serves its UI, and answers all API endpoints with the schema-correct mock mail backend; the Python register flow launches Chromium and extracts verification codes but stops because x.ai's sign-up page reports email sign-up is unavailable, so no account or SSO was produced. 16 分钟。

将你的电脑变成一台可以从手机操控的 AI agent 工作站。

测试发现：Opensteer 0.11.2 is installed as a uv global tool, all 100 unit tests pass, the CLI responds to --version (0.11.2), --help, and -c execution, and skills are installed. 1 分钟。

MCP 服务器和 Python SDK，可从自然语言生成、运行和优化 AI agent 工作流（Generative Workflow、A2A、FastAPI）。

测试发现：After mounting a mock pydaograph and a mock LLM endpoint, this container installs FlowX from the checkout, its 623-test suite passes, and the SDK, stdio MCP server and A2A HTTP server build a 4-node workflow whose generated FastAPI service returns HTTP 200 on GET / and serves /api/run-step end to end. 35 分钟。

轻量级 AI 原生工作流构建器，面向个人和小型团队：用自然语言描述你的想法，即可在可视化画布上获得可运行的工作流，并将其发布为 API。一条命令即可通过 Docker 自托管。是 Dify / n8n 的开源替代品。

测试发现：Install completes, Next.js dev server starts on port 5000 and returns 200 on /api/health, and the full test suite of 765 tests across 68 files passes cleanly. 5 分钟。

运行你自己的 agent 组织

测试发现：Rust CLI compiles and runs, Python core imports and starts a websocket server on a loopback port, WebUI builds and all 518 tests pass, 1360 of 1410 Python tests pass (7 sandbox tests need bubblewrap which requires root). 22 分钟。

用于 Claude Code 构建无瑕疵 n8n 工作流的 n8n 技能集

测试发现：n8n-skills builds 16 valid distribution zips (98 files, 15 skills), n8n-mcp v2.90.0 MCP server responds to all 7 tools correctly via stdio protocol, skills install into ~/.codex/skills/, and all pre/post-tool-use hooks are executable. 8 分钟。

面向 AGI、抗幻觉的 AI 编码运行时，基于证据、工具、记忆、agent 和验证。

测试发现：Install and build succeed CLI reports version 0 1 55 All commands render in help 6 of 13 failing test files fixed all remaining failures are pre-existing and confirmed in the original code. 74 分钟。

停止配置你的 AI 技术栈，开始使用它。一条命令即可带来一个完整的预配置 LLM 技术栈，包含数百个可供探索的服务。

测试发现：Harbor CLI v0.5.6 installed from source, frontend web app builds, all Deno-based unit tests (41 total) pass, lint self-test green (12 bash rules, 9 compose rules, 1 boost rule, 3 orchestrator checks). Docker-dependent container test matrix blocked. 16 分钟。

这个受祝福的:octocat: GitHub Action，用于将你的:package:分发文件发布到PyPI，无需令牌。

测试发现：All Python scripts (print-pkg-names, print-hash, oidc-exchange, attestations, create-docker-action, twine-upload.sh) are importable/syntactically valid; twine-upload.sh runs end-to-end with a mock PyPI-compatible HTTP server, prints debug output, uploads an sdist (.tar.gz) and a wheel (.whl), and exits 0. 5 分钟。

CodeMachine 是一个开源工具，用于将 AI 编码 agent 编排成可重复的、长时间运行的工作流。

测试发现：CodeMachine CLI v0.8.0 installs via 'bun install' (591 packages), responds to all CLI subcommands (help, version, auth, run, step, templates, agents, mcp, import, export), and launches its TUI interface. No test suite exists in the repository. 5 分钟。

一个命令安装器，用于在自有服务器上自托管 AI 技术栈：n8n、Ollama、Open WebUI、OpenClaw、Dify、Flowise、Supabase、ComfyUI、Qdrant 及 30 多种工具。基于 Docker Compose，通过 Caddy 自动配置 HTTPS，内置监控。免费的...

测试发现：The repository is structurally valid (Docker Compose YAML, shell scripts, JSON configs, Python files) but cannot install or launch because Docker is absent and cannot be installed without root privileges in this container. 13 分钟。

基于数据库的持久化 Go 工作流

测试发现：无法验证；log 显示了它停在哪里。 87 分钟。