Argusic tests whether real software actually runs 👋 Every developer knows this moment. The README says "just npm install". Two hours later you're deep in dependency conflicts, missing env vars and a model that needs a GPU nobody mentioned. Argusic exists to answer one question with evidence: Does it actually run? 🤖 What Argusic does AI agents take real software (open source repos and Hugging Face models) and install, launch and debug it exactly like a new engineer on day one. Clean, single-use machine. Pinned commit. No human help. 📊 So far: 3,163 projects tested, every run recorded. 🔍 How it works 1️⃣ Every project is pinned to an exact commit 2️⃣ An agent gets one mission on a disposable machine: Make it work 3️⃣ A failure earns a second attempt 4️⃣ Two failures settle "could not verify". Never "broken". We don't blame code we couldn't run. 🧮 How we score (out of 100) Install 30, launch or tests 30, error recovery 20, verification depth 20. A deterministic program does the scoring. No AI model grades anything, ever. "Verified" means the agent proved the output works, not just that a process started. 🧾 Proof over opinion Every published number links to its full log and terminal recording. Nothing on Argusic is a review, and no one can pay to change a result. Doubt a verdict? Open the log and check it yourself. 🔬 Methodology: Full rules, scoring and changelog argusic.com/methodology 🏆 Best software, tested: Databases, CMS, AI agent frameworks and more, ranked by whether they actually install and launch argusic.com/best 📰 Reports: Daily and weekly summaries of what ran, what failed and what it cost argusic.com/reports 🖥️ Servers: 333 GPU cloud and VPS offers from 114 providers, with verified prices and measured spin-up times argusic.com/servers 💻 Hardware: 1,300+ laptops, desktops and mini PCs for programming and AI, with estimates of which local models each can run argusic.com/hardware 🧪 Maintainer or builder? Request a test for your repo argusic.com/request 📂 Open data, free under CC BY 4.0, updated daily. Use it, cite it, build on it: 🐙 GitHub: Every verdict, score and pinned environment github.com/argusic/does-it-r… 🤗 Hugging Face: Load the dataset straight into your notebook huggingface.co/datasets/argu… 📺 Every run on video youtube.com/@argusiccom 🤝 Follow along We build Argusic in public. Follow for daily verdicts, new reports and the occasional install that fights back. Reply with a repo you want us to test next 👇 #Argusic

wireguard_webadmin: wireguard_webadmin is an intuitive web interface for WireGuard VPN management, designed to simplify the configuration and oversight of VPNs. eduardogsilva/wireguard_webadmin · JavaScript · MIT · 289 stars wireguard_webadmin tested: 11 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit f33cf97. Install to first launch: 22 min Whole session: 23 min Errors met: 11, fixed without help: 11 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "Django web application builds with venv install, migrations run successfully, all 10 tests pass (5 Django + 5 auth-gateway), and gunicorn serves the full user flow: create-first-user, login (302 to /), and /status/ dashboard returning HTTP 200 with 'WireGuard Status | wireguard-webadmin' title." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #WireguardWebadmin #Webadmintool #Wireguard #WireguardWeb #JavaScript #GitHub #Argusic

screenREC: A really simple , ad-free & minimal web based screen recorder 📹. heysagnik/screenREC · TypeScript · MIT · 290 stars screenREC tested: 3 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit f811128. Install to first launch: 3 min Whole session: 5 min Errors met: 3, fixed without help: 3 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "pnpm install succeeded (416 packages), pnpm build succeeded (2/2 apps), Next.js 16 web server answers 200 on / /record /editor, Express API answers 200 on /health and /api/health, /api/convert converts real WebM to valid MP4." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #screenREC #Collaborate #Design #Github #GoodFirstIssue #Hacktoberfest #HacktoberfestAccepted #Argusic

login-machine: AI-powered login automation. RichardHruby/login-machine · TypeScript · MIT · 292 stars login-machine tested: it runs, but only against stand-in services, never the real thing. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit e1da929. Whole session: 17 min Errors met: 4, fixed without help: 4 Argusic Score: 92/100 Depth: checked against stand-in services only The agent's own closing words: "The Next.js app builds, launches, returns HTTP 200 on / and /api/chat, and responds to all API actions (start, submit, close, error handling) using local Playwright Chromium with a mock heuristic page classifier." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #LoginMachine #AiAgents #BrowserAutomation #Browserbase #Claude #ComputerUse #Llm #Argusic

hivepaas: Self hosted PaaS solution. hivepaas/hivepaas · Go · Apache-2.0 · 292 stars hivepaas tested: 1 error, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit f7185c3. Whole session: 12 min Errors met: 1, fixed without help: 1 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "Go 1.27.2 toolchain installed, all 170+ dependency packages resolved, hivepaas binary built successfully, full test suite passes (exit 0) with no failures" The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Hivepaas #Automation #Cloud #CloudInfrastructure #Deployment #DeploymentAutomation #Devops #Argusic

cupcake: A native policy enforcement layer for AI coding agents. eqtylab/cupcake · Rust · Apache-2.0 · 297 stars cupcake tested: 2 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 27b94b8. Install to first launch: 18 min Whole session: 19 min Errors met: 2, fixed without help: 2 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "Cupcake 0.5.2 builds from source, the CLI binary runs (cupcake --version prints 0.5.2), supports 4 harnesses (Claude Code, Cursor, Factory AI, OpenCode), compiles Rego policies into WASM via OPA, evaluates hook events against builtin policies (git_pre_check, protected_paths, git_block_no_verify), and returns Allow/Deny decisions." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Cupcake #AgentSecurity #Alignment #ClaudeCode #CodingAgents #Cursor #FactoryAi #Argusic

RepoWiki: Open-source DeepWiki alternative — generate comprehensive wiki documentation for any codebase from terminal or browser. he-yufeng/RepoWiki · Python · MIT · 298 stars RepoWiki tested: nothing broke, ran first time. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 721184b. Install to first launch: 1 min Whole session: 3 min Errors met: 0, fixed without help: 0 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "All 134 tests pass, the CLI commands (scan, map, config, serve) execute successfully, the web server starts and serves HTTP 200 on /, and scan produces a complete wiki directory structure even without an LLM API key" The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #RepoWiki #Cli #Deepwiki #Documentation #Llm #Pagerank #Python #Argusic

Requester: Powerful, modern HTTP/REST client built on top of the Requests library. kylebebak/Requester · Python · MIT · 303 stars Requester tested: it runs, but only against stand-in services, never the real thing. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit e1b49b1. Install to first launch: 3 min Whole session: 8 min Errors met: 3, fixed without help: 3 Argusic Score: 92/100 Depth: checked against stand-in services only The agent's own closing words: "All 16 core unit tests pass; deps/ directory is on sys.path, Requester package is importable via symlink, and a mock HTTP server provides endpoints for requests during testing." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Requester #Api #Curl #Debug #Graphql #Http #HttpClient #Argusic

industrial_ci: Easy continuous integration repository for ROS repositories. ros-industrial/industrial_ci · Shell · Apache-2.0 · 300 stars industrial_ci tested: it would not run on a clean machine. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 125164b. Install to first launch: 4 min Whole session: 8 min Errors met: 0, fixed without help: 0 Argusic Score: 80/100 Depth: checked against stand-in services only The agent's own closing words: "The Python travis.py module correctly parses .travis.yml configurations and lists CI jobs. All shell scripts pass syntax checks. The run_travis entrypoint lists 3 jobs from the repo own .travis.yml. Full CI execution requires Docker (not available in this container)." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #IndustrialCi #Catkin #ContinuousIntegration #Gitlab #Ros #Travis #Shell #Argusic

fora: Run your own organization of agents. ImFeH2/fora · Python · Apache-2.0 · 318 stars fora tested: it runs, but only against stand-in services, never the real thing. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit f19b930. Install to first launch: 22 min Whole session: 22 min Errors met: 7, fixed without help: 6 Argusic Score: 89.14/100 Depth: checked against stand-in services only The agent's own closing words: "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)." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Fora #Agent #AgentRuntime #AgenticWorkflows #Ai #AiAgent #AiWorkspace #Argusic

parklife: 🌳 Render a Rack app (Rails/Sinatra/etc) to a static build so it can be served by Netlify, Now, GitHub Pages, S3, or any other web server. benpickles/parklife · Ruby · MIT · 308 stars parklife tested: 4 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 022b3ea. Whole session: 4 min Errors met: 4, fixed without help: 4 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "Parklife 0.9.0 builds static sites from Rack apps via CLI, all 95 specs pass, and the rack example generates correct index/foo/bar html files." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Parklife #Jamstack #Rails #Ruby #StaticSite #StaticSiteGenerator #GitHub #Argusic

ha-sun2: Home Assistant Sun2 Sensor. pnbruckner/ha-sun2 · Python · Unlicense · 309 stars ha-sun2 tested: 3 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit eef7ccf. Install to first launch: 8 min Whole session: 11 min Errors met: 3, fixed without help: 3 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "The sun2 custom component for Home Assistant is installed, imports cleanly, passes all 31 unit tests exercising sun event calculations, and is loaded at HA startup with YAML config validated." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #HaSun2 #HomeAssistant #HomeAutomation #HomeassistantCustomComponent #HomeassistantSensor #Sun #Python #Argusic

horizon: Minecraft: Java Edition 26.3 rewritten from scratch in Rust, checked against vanilla with 281 differential harnesses. ludovic111/horizon · Rust · MIT · 1,790 stars horizon tested: it runs, but only against stand-in services, never the real thing. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 4a67520. Install to first launch: 16 min Whole session: 17 min Errors met: 4, fixed without help: 4 Argusic Score: 92/100 Depth: checked against stand-in services only The agent's own closing words: "Horizon builds and all Rust crate tests pass (1074 client lib, 159 world, 104 core, 84 protocol, 50 server, 257 Python), launcher builds and tests (117/1 flaky), dedicated server binary responds to commands; GUI client requires a GPU with Vulkan to render, which this container lacks." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Horizon #DifferentialTesting #GameEngine #Gamedev #Linux #Minecraft #MinecraftJava #Argusic

xiryn: Xiryn — the open-source personal agent. SouravBeraAkaSpeed/xiryn · TypeScript · MIT · 339 stars xiryn tested: it runs, but only against stand-in services, never the real thing. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 0a55cd9. Install to first launch: 4 min Whole session: 6 min Errors met: 2, fixed without help: 2 Argusic Score: 92/100 Depth: checked against stand-in services only The agent's own closing words: "Xiryn v0.5.1 on Next.js 16.4.3 with Node.js 22.13.0: npm install, build, typecheck, lint, and format all pass; the web app starts and serves all routes (200) with a correctly initialized SQLite database (12 tables) and a mock API key in .env; Docker is unavailable in this container so agent workstation features could not be tested." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Xiryn #AiAgents #AiSdk #Anthropic #Byok #ComputerUse #Docker #Argusic

sayna: Sayna is a unified Voice Layer for AI Agents with a seemless integration to an existing agentic frameworks. SaynaAI/sayna · Rust · Apache-2.0 · 314 stars sayna tested: it runs, but only against stand-in services, never the real thing. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 2be27ae. Install to first launch: 8 min Whole session: 19 min Errors met: 6, fixed without help: 6 Argusic Score: 92/100 Depth: checked against stand-in services only The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Sayna #AiAgents #SpeechToText #TextToSpeech #VoiceAgents #VoiceAi #VoiceAiAgents #Argusic

shodh-memory: Local, LLM-free memory for AI agents. varun29ankuS/shodh-memory · Rust · Apache-2.0 · 316 stars shodh-memory tested: 3 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 8ccaf6a. Install to first launch: 21 min Whole session: 21 min Errors met: 3, fixed without help: 3 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "The Shodh-Memory v0.2.0 pre-built server binary stores and recalls memories via REST API on localhost:3031, the MCP server TypeScript test suite passes 216/216, and the Python client test suite passes 106/106." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #ShodhMemory #AgentMemory #AgenticAi #AiAgents #AiMemory #Claude #CognitiveMemory #Argusic

xcodec: AAAI 2025: Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model. zhenye234/xcodec · Python · MIT · 317 stars xcodec tested: 3 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 4a737f2. Install to first launch: 5 min Whole session: 8 min Errors met: 3, fixed without help: 3 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "X-Codec model builds, imports, runs encode/decode on real audio across all bandwidths, and produces valid WAV output" The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Xcodec #Audio #AudioCodec #Codec #Gpt #LanguageModel #Music #Argusic

platypush: A versatile and extensible platform for automation with hundreds of supported integrations. blacklight/platypush · Python · MIT · 317 stars platypush tested: 2 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 82a2334. Whole session: 6 min Errors met: 2, fixed without help: 2 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "Platypush 1.3.37 installed in a Python venv, all 120 tests pass, the CLI entry point reports help, and the HTTP backend starts and responds on port 8123." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Platypush #Automation #Bluetooth #Dashboard #HomeAutomation #Iot #Mqtt #Argusic

FrameScript: Video editing software with web frontend stacks. frame-script/FrameScript · TypeScript · MIT · 319 stars FrameScript tested: 3 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit df01b91. Install to first launch: 6 min Whole session: 10 min Errors met: 3, fixed without help: 3 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "The FrameScript project is fully installed: npm dependencies installed, studio and render Vite dev servers serve HTTP 200, Rust backend serves 6 working endpoints (healthz, is_canceled, render_log, render_audio_plan, reset), Electron v43.0.0 binary is present and responds to --version, and all build artifacts (dist/, dist-render/, dist-electron/, backend/render binaries) have been generated successfully." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #FrameScript #React #Rust #Typescript #VideoEditing #TypeScript #GitHub #Argusic

llmio: Unified LLM gateway with weighted load balancing, observability & cost tracking. atopos31/llmio · Go · MIT · 322 stars llmio tested: 5 errors, no human help, still ran. It was installed and run by an Argusic agent on a clean machine, with no human help, from the repository pinned at commit 3919f33. Install to first launch: 10 min Whole session: 16 min Errors met: 5, fixed without help: 5 Argusic Score: 100/100 Depth: checked against the real service The agent's own closing words: "The LLMIO Go server builds, the webui builds via pnpm, all 31 Go unit tests pass, and the server starts on :7070 serving the webui and responding to all API endpoints." The whole session is in this video, unedited, at the speed it actually happened. Every number above is read from that run's own log. Follow for a new repository put through this every day. #Llmio #Ai #AiGateway #Claude #ClaudeAi #ClaudeCode #Codex #Argusic