[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-835":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":22,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":31,"readmeContent":32,"aiSummary":33,"trendingCount":16,"starSnapshotCount":16,"syncStatus":34,"lastSyncTime":35,"discoverSource":36},835,"hermes-ecosystem","ksimback\u002Fhermes-ecosystem","ksimback","🗺️ Hermes Atlas — the community map of every tool, skill, and integration for Hermes Agent by Nous Research. Live at hermesatlas.com","https:\u002F\u002Fhermesatlas.com",null,"HTML",975,82,4,20,0,24,56,195,72,96.76,false,"main",true,[26,27,28,29,30],"ai-agents","ecosystem-map","hermes-agent","hermes-atlas","nous-research","2026-06-12 04:00:06","# 🗺️ Hermes Atlas\n\n**The community-curated map of every tool, skill, and integration for [Hermes Agent](https:\u002F\u002Fgithub.com\u002FNousResearch\u002Fhermes-agent) by [Nous Research](https:\u002F\u002Fnousresearch.com).**\n\n🌐 **Live site:** [hermesatlas.com](https:\u002F\u002Fhermesatlas.com)\n\n---\n\n## What is this?\n\nHermes Atlas is a living directory of the Hermes Agent ecosystem. Hermes Agent (the self-improving AI agent from Nous Research) launched in February 2026 and immediately spawned a fast-growing community of skills, plugins, integrations, deployment templates, and forks. This site is the canonical map of all of it.\n\n**Features:**\n- **80+ quality-filtered repos** across 12 categories — every project security-reviewed before inclusion\n- **Live star counts** fetched from the GitHub API and cached in Redis\n- **Sparklines and trending badges** showing growth over time\n- **RAG-powered chatbot** (\"Ask the Atlas\") that answers questions about Hermes Agent grounded in 27 research files\n- **Search, sort, filter** across the entire ecosystem\n- **Light and dark mode** with OS preference detection\n- **Mobile responsive**\n\n## How it works\n\n```\n┌─────────────────────────────────────────────────────────────┐\n│  hermesatlas.com (Vercel static + serverless)               │\n│                                                              │\n│  ┌─────────────┐  ┌──────────────┐  ┌──────────────────┐    │\n│  │ index.html  │  │ \u002Fapi\u002Fstars   │  │ \u002Fapi\u002Fchat        │    │\n│  │ (the map)   │  │ (live data)  │  │ (RAG chatbot)    │    │\n│  └─────────────┘  └──────┬───────┘  └────────┬─────────┘    │\n│                          │                    │              │\n│                          ▼                    ▼              │\n│                   ┌──────────┐         ┌────────────┐       │\n│                   │  Redis   │         │ OpenRouter │       │\n│                   │  Cloud   │         │  (Gemma 4) │       │\n│                   └────┬─────┘         └────────────┘       │\n│                        │                                     │\n│                        ▼                                     │\n│                  ┌──────────┐                                │\n│                  │ GitHub   │                                │\n│                  │ GraphQL  │                                │\n│                  └──────────┘                                │\n└─────────────────────────────────────────────────────────────┘\n```\n\n**Stack:**\n- **Frontend:** Vanilla HTML\u002FCSS\u002FJavaScript (no framework, no build step)\n- **Hosting:** Vercel (static + serverless functions)\n- **Cache:** Redis Cloud (1hr TTL on star counts, daily history snapshots)\n- **LLM:** OpenRouter with fallback chain (Gemma 4 31B → Gemma 4 26B → Gemini 3 Flash)\n- **Embeddings:** OpenAI text-embedding-3-small (computed once at build time, cached as static JSON)\n- **Retrieval:** Hybrid BM25 + cosine similarity, MMR re-ranking, conversation-aware query rewriting\n\n## Repository structure\n\n```\nhermes\u002F\n├── index.html              # The map (single-page app)\n├── api\u002F\n│   ├── stars.js            # GitHub star fetch + Redis cache\n│   ├── stars-history.js    # 30-day history for sparklines\n│   └── chat.js             # RAG pipeline with streaming\n├── lib\u002F\n│   └── redis.js            # Shared Redis client helper\n├── data\u002F\n│   ├── repos.json          # 84 quality-filtered repos (single source of truth)\n│   └── chunks.json         # Pre-computed text embeddings (~7MB, 283 chunks)\n├── scripts\u002F\n│   ├── build-chunks.js     # Splits research\u002F into chunks + embeds them\n│   └── test-rag.js         # Local RAG quality tests (27\u002F27 passing)\n├── research\u002F               # 27 source-of-truth research files\n├── repos\u002F                  # Star count data, security review, raw search results\n├── ECOSYSTEM.md            # Markdown version of the map\n├── package.json            # Just two deps: openai, redis\n└── vercel.json             # Function config + daily cron\n```\n\n## Running locally\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fksimback\u002Fhermes-ecosystem.git\ncd hermes-ecosystem\n\n# Install deps (only needed for the API endpoints + chunk builder)\nnpm install\n\n# To rebuild the chatbot's knowledge base after editing research\u002F files:\nOPENROUTER_API_KEY=sk-or-... node scripts\u002Fbuild-chunks.js\n\n# To test the RAG pipeline locally:\nOPENROUTER_API_KEY=sk-or-... node scripts\u002Ftest-rag.js\n\n# To preview the static site, just open index.html in a browser\n# (the API endpoints only work when deployed to Vercel)\n```\n\n## Environment variables (Vercel)\n\n| Variable | Purpose | Required |\n|----------|---------|----------|\n| `GITHUB_TOKEN` | Fine-grained PAT, public repos read-only | For 5000\u002Fhr rate limit (60\u002Fhr without) |\n| `OPENROUTER_API_KEY` | LLM API key for chat | Yes |\n| `REDIS_URL` | Redis Cloud connection string | Yes (for cache + history) |\n| `OPENROUTER_MODEL` | Override primary LLM model | No (default: `google\u002Fgemma-4-31b-it:free`) |\n| `OPENROUTER_FALLBACK_MODELS` | Comma-separated fallback chain | No |\n\n## Contributing\n\nFound a Hermes Agent project that should be in the map? [Open an issue](https:\u002F\u002Fgithub.com\u002Fksimback\u002Fhermes-ecosystem\u002Fissues) with the GitHub URL. Filtering criteria:\n\n- Must be specifically built for or integrated with Hermes Agent\n- Created after July 22, 2025 (Hermes repo creation date)\n- Not a personal pet project or assignment\n- Shows genuine effort and adds value to the ecosystem\n- Passes a basic security review\n\n## License\n\nSite code: MIT. Research content: CC BY 4.0. Repo descriptions and metadata are sourced from the upstream projects' own documentation.\n\n---\n\nBuilt by [@ksimback](https:\u002F\u002Fgithub.com\u002Fksimback). Not officially affiliated with Nous Research — community project celebrating their work.\n","Hermes Atlas 是由 Nous Research 创建的一个社区驱动的地图，旨在展示 Hermes Agent 生态系统中的所有工具、技能和集成。该项目通过质量筛选的80多个仓库，覆盖12个类别，并提供实时星标数量、趋势图和增长徽章等功能，确保每个项目都经过安全审查。此外，还配备了一个基于RAG技术的聊天机器人，能够回答关于Hermes Agent的问题。用户可以搜索、排序和过滤整个生态系统的内容。Hermes Atlas 采用纯HTML\u002FCSS\u002FJavaScript开发，前端无框架且无需构建步骤，后端使用Vercel进行静态加无服务器函数托管，Redis Cloud缓存数据，OpenRouter提供语言模型支持。此项目特别适合需要了解或参与到Hermes Agent相关开发工作的开发者和技术爱好者。",2,"2026-06-11 02:39:41","CREATED_QUERY"]