[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:auto-company":3},"\u003Cdiv align=\"center\">\u003Ch1>Auto Company\u003C\u002Fh1>\n\u003Cp>\u003Cstrong>A fully autonomous AI company running 24\u002F7\u003C\u002Fstrong> \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FMaxMiksa\u002Fauto-company\u002Fblob\u002FHEAD\u002FREADME-ZH.md\" rel=\"nofollow ugc noopener\">\u003Cimg alt=\"[中文说明]\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%5B%E4%B8%AD%E6%96%87%E8%AF%B4%E6%98%8E%5D-2f3640.svg\" \u002F>\u003C\u002Fa>\u003C\u002Fp>\n\u003Cp>Powered by \u003Cstrong>Agentic Workflows\u003C\u002Fstrong>, this project orchestrates 14 \u003Cstrong>Autonomous AI Agents\u003C\u002Fstrong>, each modeled after world-class experts in their domain.\nThey ideate products, make decisions, write code, deploy, and market - without human intervention.\u003C\u002Fp>\n\u003Cp>Powered by Claude Code (default) and \u003Ca href=\"https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@openai\u002Fcodex\" rel=\"nofollow ugc noopener\">Codex CLI\u003C\u002Fa> on macOS + Windows\u002FWSL, with a local dashboard on both hosts.\u003C\u002Fp>\n\u003Cp>\u003Ca href=\"#dependencies\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPlatform-macOS-blue?logo=apple&logoColor=white\" alt=\"macOS\" \u002F>\u003C\u002Fa>\n\u003Ca href=\"#windows-wsl-quick-start\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPlatform-Windows%20WSL-blue?logo=windows&logoColor=white\" alt=\"Windows WSL\" \u002F>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002F@openai\u002Fcodex\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEngine-Codex%20CLI-orange?logo=data:image\u002Fsvg%2Bxml;base64,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&logoColor=white\" alt=\"Codex CLI\" \u002F>\u003C\u002Fa>\n\u003Ca href=\"#dependencies\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FEngine-Claude%20Code-purple?logo=anthropic&logoColor=white\" alt=\"Claude Code\" \u002F>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-green?logo=opensourceinitiative&logoColor=white\" alt=\"License: MIT\" \u002F>\u003C\u002Fa>\u003C\u002Fp>\n\u003C\u002Fdiv>\u003Chr \u002F>\n\u003Ch2>Dashboard Preview\u003C\u002Fh2>\n\u003Cp>\u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002FMaxMiksa\u002Fauto-company\u002FHEAD\u002Fpresentation\u002Fdashboard-showcase.png\" alt=\"Auto Company Dashboard\" \u002F>\u003C\u002Fp>\n\u003Ch2>What Is This?\u003C\u002Fh2>\n\u003Cp>You start a loop. The AI team wakes up, reads shared consensus memory, decides what to do, forms a 3-5 person squad, executes, updates consensus memory, then sleeps briefly. Then it repeats.\u003C\u002Fp>\n\u003Cpre>\u003Ccode>daemon (launchd \u002F systemd --user, auto-restart on crash)\n  └── scripts\u002Fcore\u002Fauto-loop.sh (continuous loop)\n        ├── reads PROMPT.md + consensus.md\n        ├── LLM CLI call (Codex CLI \u002F Claude Code)\n        │   ├── reads CLAUDE.md (charter + guardrails)\n        │   ├── reads .claude\u002Fskills\u002Fteam\u002FSKILL.md (teaming method)\n        │   ├── forms an Agent Team (3-5 agents)\n        │   ├── executes: research, coding, deploy, marketing\n        │   └── updates memories\u002Fconsensus.md (handoff baton)\n        ├── failure handling: rate-limit wait \u002F circuit breaker \u002F consensus rollback\n        └── sleep -> next cycle\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Each cycle is an independent CLI call. \u003Ccode>memories\u002Fconsensus.md\u003C\u002Fcode> is the only cross-cycle state.\u003C\u002Fp>\n\u003Ch2>Where To Start (By Platform)\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>Windows users: start from [Wind\u003C\u002Fli>\n\u003C\u002Ful>\n",1784240406338]