[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-75415":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":15,"stars7d":16,"stars30d":17,"stars90d":14,"forks30d":14,"starsTrendScore":18,"compositeScore":19,"rankGlobal":9,"rankLanguage":9,"license":9,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":9,"pushedAt":9,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":14,"starSnapshotCount":14,"syncStatus":27,"lastSyncTime":28,"discoverSource":29},75415,"oh-my-hermes","Salomondiei08\u002Foh-my-hermes","Salomondiei08","An opinionated workflow layer for building, shipping, and operating apps with Hermes Agent",null,"Shell",485,44,106,0,9,61,381,27,89.96,false,"main",true,[],"2026-06-12 04:01:18","\u003Cimg src=\"banner.png\" alt=\"Oh My Hermes\" width=\"100%\" \u002F>\n\n# Oh My Hermes\n\n[![Stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsalomondiei08\u002Foh-my-hermes?style=flat-square)](https:\u002F\u002Fgithub.com\u002Fsalomondiei08\u002Foh-my-hermes\u002Fstargazers)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-blue?style=flat-square)](LICENSE)\n[![Hermes](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHermes-v0.13%2B-orange?style=flat-square)](https:\u002F\u002Fhermes-agent.nousresearch.com)\n[![Skills](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fskills-23-brightgreen?style=flat-square)](#skills)\n[![Agents](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fagents-6-blue?style=flat-square)](#agents)\n\n**An opinionated workflow layer for building, shipping, and operating apps — delivered directly to Hermes.**\n\nLike Oh My Zsh is to Zsh. You install it once and Hermes becomes genuinely useful for real software projects. Not a chatbot wrapper. Not a prompt pack. A curated set of skills that Hermes loads and runs autonomously — on your VPS, on your laptop, wherever Hermes lives.\n\n> **AI agent?** See [INSTALL_FOR_AGENTS.md](INSTALL_FOR_AGENTS.md) for the one-command install protocol.\n\n---\n\n## Use with an AI agent\n\nIf you work with Claude, Cursor, Copilot, or any coding assistant, paste this directly into the chat:\n\n**To install Oh My Hermes on a project:**\n\n```\nInstall Oh My Hermes on this project.\n\n1. Clone the repo: git clone https:\u002F\u002Fgithub.com\u002Fsalomondiei08\u002Foh-my-hermes \u002Ftmp\u002Foh-my-hermes\n2. Run the installer: bash \u002Ftmp\u002Foh-my-hermes\u002Finstall.sh\n3. Bootstrap this project: bash \u002Ftmp\u002Foh-my-hermes\u002Fscripts\u002Fbootstrap.sh\n4. Verify everything installed: bash \u002Ftmp\u002Foh-my-hermes\u002Fscripts\u002Fverify.sh\n\nThen tell me what's missing and what I need to fill in.\n```\n\n**To set up the full autonomous CTO loop:**\n\n```\nSet up the Oh My Hermes CTO loop for this project.\n\nFollow the instructions in: https:\u002F\u002Fraw.githubusercontent.com\u002Fsalomondiei08\u002Foh-my-hermes\u002Fmain\u002FINSTALL_FOR_AGENTS.md\n\nI will provide the following when you ask:\n- My GitHub repo (owner\u002Frepo)\n- A GitHub fine-grained token (I'll create one if you explain how)\n- My production URL (if deployed)\n\nWalk me through each step one at a time.\n```\n\n**To start a new app from scratch:**\n\n```\nUse Oh My Hermes to start a new app.\n\nOh My Hermes is installed at ~\u002F.hermes\u002Fskills\u002F. Load the following skills in order:\n1. clarify-requirements — ask me the 7 questions and save my answers\n2. product-brief — generate the brief from my answers\n3. choose-engine — decide how to implement it\n4. implement — build it using the right engine\n\nStart with clarify-requirements now.\n```\n\n---\n\n## The core idea\n\nHermes is the operator. It talks to you. It remembers. It builds. It deploys. It monitors. It notifies.\n\nYou describe what you want in plain language — on Telegram, Slack, Discord, your terminal, wherever you have Hermes configured. Hermes loads the right skill and runs the workflow. Claude Code and Codex are optional engines Hermes can invoke when a task needs deep file editing — but Hermes handles the orchestration, the ops, the memory, and the lifecycle on its own.\n\n**Hermes does not need Claude Code or Codex to be useful.** Those are optional. Hermes itself has a terminal backend and can write, edit, and run code directly.\n\n---\n\n## How it all fits together\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│                        YOU (founder)                            │\n│         Telegram · Slack · Discord · WhatsApp · terminal        │\n└──────────────────────────┬──────────────────────────────────────┘\n                           │  plain-language messages\n                           ▼\n┌─────────────────────────────────────────────────────────────────┐\n│                   HERMES  (VPS \u002F local, 24\u002F7)                   │\n│                                                                 │\n│  ┌──────────┐   routes to   ┌─────────────────────────────────┐ │\n│  │ Gateway  │ ────────────▶ │        CTO Agent                │ │\n│  │(Telegram │               │  monitors kanban, orchestrates  │ │\n│  │ \u002FSlack…) │               └───────────┬─────────────────────┘ │\n│  └──────────┘                           │  spawns sub-agents    │\n│                      ┌──────────────────┼──────────────┐        │\n│                      ▼                  ▼              ▼        │\n│               ┌────────────┐  ┌──────────────┐  ┌──────────┐   │\n│               │  PM · Dev  │  │ QA · Security│  │   Ops    │   │\n│               │ triage     │  │ review · scan│  │ deploy   │   │\n│               │ implement  │  │ approve      │  │ monitor  │   │\n│               └─────┬──────┘  └──────┬───────┘  └────┬─────┘   │\n│                     │                │               │          │\n│              ┌──────▼────────────────▼───────────────▼──────┐   │\n│              │              Hermes Kanban                    │   │\n│              │   Backlog → In Progress → Review → Done       │   │\n│              └───────────────────────────────────────────────┘   │\n│                                                                 │\n│  Persistent memory · 23 skills · 5 workflows · cron jobs        │\n└──────┬──────────────────────────────────────────────────────────┘\n       │\n       ├──▶  Claude Code  (complex multi-file coding)\n       ├──▶  Codex        (quick single-file fixes)\n       ├──▶  Vercel       (hosting + preview URLs)\n       ├──▶  Supabase     (database + auth + migrations)\n       ├──▶  GitHub       (issues, PRs, merge)\n       └──▶  Sentry \u002F Uptime Kuma  (monitoring)\n```\n\n---\n\n## Full project lifecycle\n\nFrom idea to running production app, Hermes handles each stage:\n\n```\nIDEA\n  ↓  onboarding            ← Bot asks everything in chat, configures the loop\n  ↓  clarify-requirements  ← Hermes asks 7 questions, saves answers to memory\n  ↓  product-brief         ← Writes PRODUCT_BRIEF.md from requirements\n  ↓  design-handoff        ← Converts your design notes to an implementation spec\n  ↓  choose-engine         ← Decides: Hermes \u002F Claude Code \u002F Codex\n  ↓  implement             ← Builds it — surgical changes, no secrets committed\n  ↓  deploy-to-vercel      ← Pre-deploy checks → deploy → captures URL\n  ↓  connect-supabase      ← Links DB, pushes migrations, sets env vars\n  ↓  setup-monitoring      ← Sentry + Uptime Kuma\n  ↓  post-deploy-followup  ← Health check → log → notify you\n  ↓\nRUNNING APP\n  ↓  auto-issue-triage     ← Every hour: scores issues, picks top priority\n  ↓  implement + PR        ← Dev Agent builds, Security Agent reviews\n  ↓  await-merge-approval  ← Sends you YES\u002FNO message\n  ↓  YOU reply YES         ← Merges, deploys, health-checks, confirms\n  ↓\nREPEAT — Hermes watches it, you approve changes\n```\n\n---\n\n## The autonomous CTO loop\n\nOnce configured, this runs every hour without you touching anything:\n\n```\nGitHub issue opens\n       ↓\n  PM Agent scores & triages  →  kanban: Backlog\n       ↓\n  Dev Agent implements  →  kanban: In Progress\n       ↓\n  Security Agent: secret scan + OWASP check + CVE check\n       ↓\n  QA Agent: build check + health check + plain-English summary\n       ↓\n  YOU get a message on Telegram:\n  ────────────────────────────────────────\n  PR #12 — Fix login redirect\n\n  What changed: Users who hadn't verified their email were\n  sent to a broken page. They now see \"Check your inbox.\"\n\n  Build: passing  |  Preview: healthy (180ms)  |  No secrets found\n  Preview: https:\u002F\u002Fmyapp-fix-login.vercel.app\n\n  Reply YES to ship. Reply NO and tell me why.\n  ────────────────────────────────────────\n       ↓\n  YES → merges, deploys, health-checks, confirms live URL to you\n  NO  → Dev Agent iterates on your feedback\n```\n\n---\n\n## Get started\n\n**Step 1 — Install Hermes Agent**\n\nFollow the [Hermes quickstart](https:\u002F\u002Fhermes-agent.nousresearch.com\u002Fdocs\u002Fgetting-started\u002Fquickstart). At the end you have a bot you can message on Telegram (or Slack, Discord, WhatsApp).\n\n**Step 2 — Install Oh My Hermes**\n\n```bash\ncurl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002Fsalomondiei08\u002Foh-my-hermes\u002Fmain\u002Finstall.sh | bash\n```\n\n**Step 3 — Message your bot**\n\n```\nset up the CTO loop\n```\n\nThe bot will ask for your GitHub repo, walk you through creating a token step by step, ask for your production URL, and configure everything. No terminal needed after Step 2.\n\n**What you unlock at each step:**\n\n| Step | What to do | What you get |\n|---|---|---|\n| 1 | Install Hermes + connect Telegram | A bot you can message |\n| 2 | Run install.sh | 23 skills and 5 workflows loaded |\n| 3 | Message: \"set up the CTO loop\" | Bot guides the full setup in chat |\n| 4 | Bot asks for GitHub token | Issues monitored, PRs managed |\n| 5 | Bot asks for production URL | Health checks every 15 min |\n| 6 | `\u002Fgoal` command | Agent stays focused across long sessions |\n| — | Autonomous from here | Hourly triage, daily report, weekly security scan |\n\n---\n\n## Skills\n\n| Skill | What Hermes does |\n|---|---|\n| `onboarding` | Guides full setup in chat — no terminal, no manual config |\n| `clarify-requirements` | Asks 7 structured questions, saves answers to memory |\n| `product-brief` | Generates a product brief, writes PRODUCT_BRIEF.md |\n| `design-handoff` | Converts design notes to an implementation spec |\n| `create-skill` | Creates a new skill in the correct format (meta-skill) |\n| `choose-engine` | Routes tasks to Hermes, Claude Code, or Codex |\n| `implement-with-claude-code` | Scaffolds Claude Code with full context + scope constraints |\n| `implement-with-codex` | Scaffolds Codex for targeted single-file fixes |\n| `deploy-to-vercel` | Pre-deploy checks → deploy → capture URL |\n| `connect-supabase` | Links Supabase, pushes migrations, sets Vercel env vars |\n| `setup-monitoring` | Configures Sentry + Uptime Kuma |\n| `health-check` | Calls `\u002Fapi\u002Fhealth`, validates response, checks Supabase + Vercel logs |\n| `send-notification` | Sends Slack webhook with deployment or status info |\n| `post-deploy-followup` | Health check + deployment log + notification + summary |\n| `manage-github-issues` | Triage, create, label, assign, and close GitHub issues |\n| `create-github-pr` | Creates PR with secret scan before opening |\n| `auto-issue-triage` | Hourly: scores open issues, picks top priority, starts work |\n| `review-github-pr` | Reviews PR diff, runs checks, writes plain-English summary |\n| `security-review` | Secret scan + OWASP check + CVE audit + weekly supply chain |\n| `await-merge-approval` | Sends YES\u002FNO message to founder, merges or iterates |\n| `kanban-task` | Creates and updates Hermes kanban cards at every stage |\n| `cto-status-report` | Daily morning report: what's in progress, done, blocked |\n| `backup-hermes-data` | Tarballs `~\u002F.hermes\u002F` to S3, Dropbox, or local |\n\n---\n\n## Agents\n\n| Agent | Role | Kanban ownership |\n|---|---|---|\n| **CTO** | Orchestrates all agents, monitors kanban, reports to you daily | All columns |\n| **PM** | Triages GitHub issues, writes tickets, prioritizes backlog | Backlog |\n| **Dev** | Implements tickets, picks the right engine, creates PRs | In Progress |\n| **Security** | Scans every PR for secrets, OWASP issues, and CVEs | Between Dev and QA |\n| **QA** | Reviews PRs, runs health checks, writes founder summary | Review |\n| **Ops** | Deploys, monitors production, handles incidents | Done + monitoring |\n\nRole definitions live in `agents\u002F`. Running `scripts\u002Fsetup-cto.sh` (or messaging \"set up the CTO loop\") creates the actual Hermes profiles and makes the agents active.\n\n---\n\n## Workflow examples\n\n**Start a new project:**\n```\nyou: start a new app\nhermes: What problem does this solve? Who experiences it?\nyou: [answer]\nhermes: [6 more questions…]\nhermes: Requirements saved. Generating product brief…\nhermes: Brief written to PRODUCT_BRIEF.md. Ready to implement or do design first?\n```\n\n**Deploy after implementing:**\n```\nyou: deploy this to Vercel\nhermes: Running pre-deploy checklist…\nhermes: Deploying… done. URL: https:\u002F\u002Fmyapp.vercel.app\nhermes: Health check: PASS (200ms)\nhermes: Notification sent to Slack.\n```\n\n**Quick fix:**\n```\nyou: fix the auth redirect bug in src\u002Fmiddleware.ts\nhermes: Loading context… routing to Codex (single-file fix)\nhermes: Done. Typecheck passes. Creating PR…\nhermes: PR #14 ready — reply YES to ship.\n```\n\n**Steer mid-session (Hermes v0.13+):**\n```\nyou: \u002Fsteer prioritize the payment bug above everything else\nhermes: Understood. Switching Dev Agent to issue #38.\n```\n\n---\n\n## Default stack\n\n| Layer | Default | Alternative |\n|---|---|---|\n| Frontend \u002F full-stack | Vercel | Railway, Render |\n| Database | Supabase PostgreSQL | PlanetScale, Neon |\n| Auth | Supabase Auth | Clerk, Auth.js |\n| Error tracking | Sentry | LogRocket |\n| Uptime monitoring | Uptime Kuma | Better Uptime |\n| Notifications | Slack webhook | Telegram, Email |\n\nAll pluggable. Each skill documents how to substitute.\n\n---\n\n## Running on a VPS\n\nThe intended setup for production use — Hermes runs 24\u002F7, crons fire automatically, you interact from your phone:\n\n```bash\n# On a $5\u002Fmonth VPS (Ubuntu 22.04+)\ncurl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002FNousResearch\u002Fhermes-agent\u002Fmain\u002Fscripts\u002Finstall.sh | bash\nhermes model        # choose your provider (Anthropic, OpenAI, etc.)\nhermes gateway setup && hermes gateway start   # connect Telegram or Slack\n\n# Then install Oh My Hermes\ncurl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002Fsalomondiei08\u002Foh-my-hermes\u002Fmain\u002Finstall.sh | bash\n\n# Message your bot: \"set up the CTO loop\"\n```\n\nFor Docker:\n```bash\ndocker run -d --restart=always \\\n  -v hermes-data:\u002Froot\u002F.hermes \\\n  nousresearch\u002Fhermes-agent\n```\n\n---\n\n## Installation scripts\n\n| Script | What it does |\n|---|---|\n| `install.sh` | Installs all skills, workflows, and agent definitions |\n| `scripts\u002Fbootstrap.sh` | Creates `AGENTS.md`, `.env.example`, health endpoint in a project |\n| `scripts\u002Fsetup-cto.sh` | Creates profiles, initializes kanban, schedules crons |\n| `scripts\u002Fverify.sh` | Checks everything is installed correctly |\n| `scripts\u002Funinstall.sh` | Removes all Oh My Hermes files from `~\u002F.hermes\u002F` |\n\n---\n\n## Optional: GBrain memory backbone\n\n[GBrain](https:\u002F\u002Fgithub.com\u002Fgarrytan\u002Fgbrain) gives Hermes a richer, self-updating knowledge graph — people, companies, decisions, deployment history — queryable across sessions.\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fgarrytan\u002Fgbrain.git ~\u002Fgbrain && cd ~\u002Fgbrain\ncurl -fsSL https:\u002F\u002Fbun.sh\u002Finstall | bash && export PATH=\"$HOME\u002F.bun\u002Fbin:$PATH\"\nbun install && bun link && gbrain init\n```\n\nDo not use `npm install -g gbrain` — a squatter package exists on npm under that name.\n\n---\n\n## Architecture\n\n```\noh-my-hermes\u002F\n├── skills\u002F          ← 23 skill files → ~\u002F.hermes\u002Fskills\u002F\n├── workflows\u002F       ← 5 workflow files → ~\u002F.hermes\u002Fworkflows\u002F\n├── agents\u002F          ← 6 agent role definitions → ~\u002F.hermes\u002Fagents\u002F\n├── templates\u002F       ← AGENTS.md template, .env example, health endpoint\n├── scripts\u002F         ← install, bootstrap, verify, setup-cto, uninstall\n└── docs\u002F            ← Full documentation\n```\n\nSee [docs\u002Farchitecture.md](docs\u002Farchitecture.md) for detail.\n\n---\n\n## Roadmap\n\n**V1 — current**\n23 skills, 6 agents, 5 workflows, chat-guided onboarding, security agent, Karpathy code principles, one-command CTO setup, Vercel + Supabase + GitHub ops.\n\n**V2 — planned**\nRollback skill, staging → production promotion workflow, incident creation, post-deploy automated tests.\n\n**V3 — planned**\nMulti-service orchestration, more example apps, hosted setup wizard.\n\n---\n\n## Star history\n\n[![Star History Chart](https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=salomondiei08\u002Foh-my-hermes&type=Date)](https:\u002F\u002Fstar-history.com\u002F#salomondiei08\u002Foh-my-hermes&Date)\n\n---\n\n## Contributing\n\nRead [docs\u002Farchitecture.md](docs\u002Farchitecture.md) before proposing features. Open issues for wrong or missing skills, bugs in scripts, or Hermes improvement proposals.\n\n---\n\n## License\n\nMIT\n","Oh My Hermes 是一个用于构建、部署和运维应用程序的工作流层，专为Hermes Agent设计。该项目通过提供一系列精心策划的技能集，使得Hermes能够在VPS、笔记本电脑等环境中自主运行应用开发流程，包括需求澄清、产品简报生成、技术选型及实现等核心功能。这些技能被Hermes加载后能够自动执行，极大地简化了从项目启动到持续运营的过程。特别适合那些希望利用AI助手（如Claude, Cursor, Copilot）来提高软件开发效率的开发者或团队使用，在保证高效的同时也保持了良好的可操作性和灵活性。",2,"2026-06-11 03:52:41","CREATED_QUERY"]