[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"tool-related-zh-harish-gaggar-mcp-server-gateway":3},{"items":4,"total":98,"page":99,"pageSize":100},[5,33,50,66,81],{"slug":6,"name":7,"categorySlug":8,"summary":9,"description":10,"categoryName":11,"sourceUrl":12,"homepageUrl":-1,"downloadUrl":13,"cliCommand":-1,"primaryLanguage":-1,"license":14,"stars":15,"forks":16,"trendScore":17,"qualityScore":18,"supportedClients":19,"clients":22,"tags":30,"installCommand":-1,"configSnippet":-1,"useCases":31,"updatedAt":32},"headroomlabs-ai-headroom","headroomlabs-ai\u002Fheadroom","devtools","Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.","██╗ ██╗███████╗ █████╗ ██████╗ ██████╗ ██████╗ ██████╗ ███╗ ███╗ ██║ ██║██╔════╝██╔══██╗██╔══██╗██╔══██╗██╔═══██╗██╔═══██╗████╗ ████║ ███████║█████╗ ███████║██║ ██║██████╔╝██║ ██║██║ ██║██╔████╔██║ ██╔══██║██╔══╝ ██╔══██║██║ ██║██╔══██╗██║ ██║██║ ██║██║╚██╔╝██║ ██║ ██║███████╗██║ ██║██████╔╝██║ ██║╚██████╔╝╚██████╔╝██║ ╚═╝ ██║ ╚═╝ ╚═╝╚══════╝╚═╝ ╚═╝╚═════╝ ╚═╝ ╚═╝ ╚═════╝ ╚═════╝ ╚═╝ ╚═╝ The context compression layer for AI agents\n\n60–95% fewer tokens (for JSON data), 15-20% fewer tokens (for coding agents) · library · proxy · MCP · content-aware compressors · local-first · reversible\n\nDocs · Install · Proof · Agents · Discord · llms.txt\n\nAI agents \u002F LLMs: read \u002Fllms.txt here, or fetch the live index \u002F full docs blob .\n\nHeadroom compresses everything your AI agent reads — tool outputs, logs, RAG chunks, files, and conversation history — before it reaches the LLM. Same answers, fraction of the tokens.\n\nLive: 10,144 → 1,260 tokens — same FATAL found.","开发工具","https:\u002F\u002Fgithub.com\u002Fheadroomlabs-ai\u002Fheadroom","https:\u002F\u002Fgithub.com\u002Fheadroomlabs-ai\u002Fheadroom\u002Farchive\u002FHEAD.zip","Apache-2.0",75300,5843,100,81,[20,21],"claude-desktop","cursor",[23,27],{"slug":20,"name":24,"summary":25,"count":26},"Claude Desktop","支持本地 MCP 配置的桌面客户端。",0,{"slug":21,"name":28,"summary":29,"count":26},"Cursor","常与开发类 MCP Server 搭配使用的 AI 代码编辑器。",[8,14],[],"2026-08-30T08:28:44.109415",{"slug":34,"name":35,"categorySlug":8,"summary":36,"description":36,"categoryName":11,"sourceUrl":37,"homepageUrl":-1,"downloadUrl":38,"cliCommand":-1,"primaryLanguage":-1,"license":39,"stars":40,"forks":41,"trendScore":17,"qualityScore":42,"supportedClients":43,"clients":44,"tags":47,"installCommand":-1,"configSnippet":-1,"useCases":48,"updatedAt":49},"sickn33-agentic-awesome-skills","sickn33\u002Fagentic-awesome-skills","Agentic Awesome Skills is a library of 2,689+ installable SKILL.md playbooks. AAS Core helps Codex or Claude search the complete local catalog, record the skills the agent chooses, and preview a plan...","https:\u002F\u002Fgithub.com\u002Fsickn33\u002Fagentic-awesome-skills","https:\u002F\u002Fgithub.com\u002Fsickn33\u002Fagentic-awesome-skills\u002Farchive\u002FHEAD.zip","MIT",47563,6935,80,[20,21],[45,46],{"slug":20,"name":24,"summary":25,"count":26},{"slug":21,"name":28,"summary":29,"count":26},[8,39],[],"2026-07-19T02:10:25.868557",{"slug":51,"name":52,"categorySlug":8,"summary":53,"description":53,"categoryName":11,"sourceUrl":54,"homepageUrl":-1,"downloadUrl":55,"cliCommand":-1,"primaryLanguage":-1,"license":39,"stars":56,"forks":57,"trendScore":17,"qualityScore":58,"supportedClients":59,"clients":60,"tags":63,"installCommand":-1,"configSnippet":-1,"useCases":64,"updatedAt":65},"punkpeye-awesome-mcp-servers","punkpeye\u002Fawesome-mcp-servers","ray.run – from idea to a production-grade MCP server in under a minute! 🦜","https:\u002F\u002Fgithub.com\u002Fpunkpeye\u002Fawesome-mcp-servers","https:\u002F\u002Fgithub.com\u002Fpunkpeye\u002Fawesome-mcp-servers\u002Farchive\u002FHEAD.zip",96308,17566,54,[20,21],[61,62],{"slug":20,"name":24,"summary":25,"count":26},{"slug":21,"name":28,"summary":29,"count":26},[8,39],[],"2026-08-14T05:27:24.263802",{"slug":67,"name":68,"categorySlug":8,"summary":69,"description":69,"categoryName":11,"sourceUrl":70,"homepageUrl":-1,"downloadUrl":71,"cliCommand":-1,"primaryLanguage":-1,"license":39,"stars":72,"forks":73,"trendScore":17,"qualityScore":42,"supportedClients":74,"clients":75,"tags":78,"installCommand":-1,"configSnippet":-1,"useCases":79,"updatedAt":80},"hkuds-nanobot","hkuds\u002Fnanobot","English | 简体中文 | 繁體中文 | Español | Français | Bahasa Indonesia | 日本語 | 한국어 | Русский | Tiếng Việt Discord · X · WeChat \u002F Feishu","https:\u002F\u002Fgithub.com\u002Fhkuds\u002Fnanobot","https:\u002F\u002Fgithub.com\u002Fhkuds\u002Fnanobot\u002Farchive\u002FHEAD.zip",49007,8629,[20,21],[76,77],{"slug":20,"name":24,"summary":25,"count":26},{"slug":21,"name":28,"summary":29,"count":26},[8,39],[],"2026-08-19T05:48:15.579337",{"slug":82,"name":83,"categorySlug":8,"summary":84,"description":85,"categoryName":11,"sourceUrl":86,"homepageUrl":-1,"downloadUrl":87,"cliCommand":-1,"primaryLanguage":-1,"license":14,"stars":88,"forks":89,"trendScore":17,"qualityScore":90,"supportedClients":91,"clients":92,"tags":95,"installCommand":-1,"configSnippet":-1,"useCases":96,"updatedAt":97},"zylon-ai-private-gpt","zylon-ai\u002Fprivate-gpt","Complete API layer for private AI applications on local models: RAG, skills, tools, MCP, text-to-sql, and more. Works with any OpenAI-compatible inference server.","Running a model locally is only the first step. To build useful AI applications you need a set of higher-level building blocks. PrivateGPT provides that layer as an open-source API following the Claude API model — so you can build private AI products without rebuilding the same backend primitives from scratch, and without depending on cloud APIs.\n\nProduction-tested: PrivateGPT powers Zylon, the on-premise AI platform providing Private AI to enterprises across the globe.\n\nPrivateGPT does run models itself. It connects to any OpenAI-compatible inference server via OPENAI_API_BASE. If it implements \u002Fv1\u002Fchat\u002Fcompletions and \u002Fv1\u002Fmodels, it works.\n\nPrivateGPT ships a built-in workbench UI for testing and demos, available at \u002Fui. The API is the actual product.","https:\u002F\u002Fgithub.com\u002Fzylon-ai\u002Fprivate-gpt","https:\u002F\u002Fgithub.com\u002Fzylon-ai\u002Fprivate-gpt\u002Farchive\u002FHEAD.zip",57575,7619,74,[20,21],[93,94],{"slug":20,"name":24,"summary":25,"count":26},{"slug":21,"name":28,"summary":29,"count":26},[8,14],[],"2026-08-30T08:30:29.511657",5512,1,5]