[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"global-settings":3,"blog-post-why-87-of-your-prompt-isnt-your-prompt":181,"markdown-2761-qigiuw":236},{"data":4},{"id":5,"documentId":6,"siteName":7,"address":8,"contactEmail":9,"copyrightText":10,"hubspotPortalId":11,"hubspotNewsletterFormId":12,"createdAt":13,"updatedAt":14,"publishedAt":15,"locale":12,"hubspotJobApplicationFormId":12,"hubspotDemoRequestFormId":12,"hubspotContactFormId":16,"defaultOgImageAlt":7,"logo":17,"logoDark":31,"defaultOgImage":39,"headerCtas":84,"legalLinks":95,"socialLinks":105,"headerNav":111,"footerColumns":168},4,"t4ysbj7xqjxe9vwpng6wvs9y","Hyground","Versmannstraße 2, 20457 Hamburg, Germany","contact@hyground.ai","© 2026 Hyground. All rights reserved.","145901011",null,"2026-04-22T08:43:29.438Z","2026-05-15T17:17:10.042Z","2026-05-15T17:17:09.865Z","496682dd-c91c-463f-b8f5-30e9679f415e",{"id":18,"documentId":19,"name":20,"alternativeText":12,"caption":12,"focalPoint":12,"width":21,"height":22,"formats":12,"hash":23,"ext":24,"mime":25,"size":26,"url":27,"previewUrl":12,"provider":28,"provider_metadata":12,"folderPath":29,"createdAt":30,"updatedAt":30,"publishedAt":30,"locale":12},69,"l1zkkv2k4anah2ethspw4kk4","logo.svg",508,103,"logo_230b5ea131",".svg","image\u002Fsvg+xml",9.8,"\u002Fuploads\u002Flogo_230b5ea131.svg","local","\u002F","2026-05-04T13:52:45.374Z",{"id":32,"documentId":33,"name":34,"alternativeText":12,"caption":12,"focalPoint":12,"width":21,"height":22,"formats":12,"hash":35,"ext":24,"mime":25,"size":36,"url":37,"previewUrl":12,"provider":28,"provider_metadata":12,"folderPath":29,"createdAt":30,"updatedAt":30,"publishedAt":38,"locale":12},70,"n75laf0qlq0cjo1ksgxafb4i","logo-dark.svg","logo_dark_793a0f8cb9",9.81,"\u002Fuploads\u002Flogo_dark_793a0f8cb9.svg","2026-05-04T13:52:45.375Z",{"id":40,"documentId":41,"name":42,"alternativeText":12,"caption":12,"focalPoint":12,"width":43,"height":44,"formats":45,"hash":80,"ext":47,"mime":50,"size":81,"url":82,"previewUrl":12,"provider":28,"provider_metadata":12,"folderPath":29,"createdAt":83,"updatedAt":83,"publishedAt":83,"locale":12},162,"wo4i957bepocr78jgqrcswz6","hyground-og-img.webp",1200,630,{"large":46,"small":56,"medium":64,"thumbnail":72},{"ext":47,"url":48,"hash":49,"mime":50,"name":51,"path":12,"size":52,"width":53,"height":54,"sizeInBytes":55},".webp","\u002Fuploads\u002Flarge_hyground_og_img_f5c8198dfa.webp","large_hyground_og_img_f5c8198dfa","image\u002Fwebp","large_hyground-og-img.webp",9.23,1000,525,9226,{"ext":47,"url":57,"hash":58,"mime":50,"name":59,"path":12,"size":60,"width":61,"height":62,"sizeInBytes":63},"\u002Fuploads\u002Fsmall_hyground_og_img_f5c8198dfa.webp","small_hyground_og_img_f5c8198dfa","small_hyground-og-img.webp",4.71,500,262,4710,{"ext":47,"url":65,"hash":66,"mime":50,"name":67,"path":12,"size":68,"width":69,"height":70,"sizeInBytes":71},"\u002Fuploads\u002Fmedium_hyground_og_img_f5c8198dfa.webp","medium_hyground_og_img_f5c8198dfa","medium_hyground-og-img.webp",6.84,750,394,6844,{"ext":47,"url":73,"hash":74,"mime":50,"name":75,"path":12,"size":76,"width":77,"height":78,"sizeInBytes":79},"\u002Fuploads\u002Fthumbnail_hyground_og_img_f5c8198dfa.webp","thumbnail_hyground_og_img_f5c8198dfa","thumbnail_hyground-og-img.webp",2.29,245,129,2288,"hyground_og_img_f5c8198dfa",11.92,"\u002Fuploads\u002Fhyground_og_img_f5c8198dfa.webp","2026-05-12T07:30:09.091Z",[85,90],{"id":86,"label":87,"url":88,"variant":89},56,"Try our Sandbox","\u002Ftry-hyground-sandbox","ghost",{"id":91,"label":92,"url":93,"variant":94},55,"Book a demo","\u002Fbook-demo","primary",[96,101],{"id":97,"label":98,"url":99,"external":100},46,"Privacy Policy","\u002Fprivacy-policy",false,{"id":102,"label":103,"url":104,"external":100},47,"Imprint","\u002Fimprint",[106],{"id":107,"label":108,"url":109,"external":110},48,"LinkedIn","https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Fhyground",true,[112,116,142,154,159,164],{"id":113,"label":114,"url":29,"external":100,"children":115},1,"Home",[],{"id":117,"label":118,"url":12,"external":100,"children":119},5,"Product",[120,123,127,131,135,139],{"id":91,"label":121,"url":122,"external":100},"Overview","\u002Fproduct-overview",{"id":124,"label":125,"url":126,"external":100},54,"Integrations","\u002Fproduct-integrations",{"id":128,"label":129,"url":130,"external":100},49,"Skills","\u002Fproduct-skills",{"id":132,"label":133,"url":134,"external":100},50,"Scheduling","\u002Fproduct-scheduling",{"id":136,"label":137,"url":138,"external":100},51,"Triggers","\u002Fproduct-triggers",{"id":86,"label":140,"url":141,"external":100},"Security","\u002Fproduct-security",{"id":143,"label":144,"url":12,"external":100,"children":145},6,"Use Cases",[146,150],{"id":147,"label":148,"url":149,"external":100},52,"Incident Investigation","\u002Fuse-cases-incident-investigation",{"id":151,"label":152,"url":153,"external":100},53,"Workflow Automation","\u002Fuse-cases-workflow-automation",{"id":155,"label":156,"url":157,"external":100,"children":158},2," Blog","\u002Fblog",[],{"id":160,"label":161,"url":162,"external":100,"children":163},3,"Company","\u002Fcompany",[],{"id":5,"label":165,"url":166,"external":100,"children":167},"Contact","\u002Fcontact",[],[169],{"id":5,"heading":170,"links":171},"Site",[172,174,177,179],{"id":173,"label":114,"url":29,"external":100},41,{"id":175,"label":176,"url":157,"external":100},42,"Blog",{"id":178,"label":165,"url":166,"external":100},44,{"id":180,"label":92,"url":93,"external":100},45,{"data":182,"meta":233},[183],{"id":184,"documentId":185,"title":186,"slug":187,"excerpt":188,"body":189,"coverImageUrl":12,"readingMinutes":143,"tags":190,"publishedDate":193,"createdAt":194,"updatedAt":195,"publishedAt":196,"coverImage":197,"categories":232},72,"imq36sly7qfch0qxqbyhc928","Why 87% of Your Prompt Isn't Your Prompt","why-87-of-your-prompt-isnt-your-prompt","Loading every available tool definition upfront causes significant performance degradation and wastes the model's limited attention budget.","When OpenAI introduced function calling in June 2023, it felt like the missing piece for building useful AI agents. Finally, LLMs could interact with the real world. But anyone who shipped production systems quickly learned the truth: it was finicky. You had to manage the tool call loop yourself, handle errors gracefully, and hope the model picked the right function from your carefully crafted definitions.\n\nThen came MCP.\n\nIn November 2024, Anthropic open-sourced the Model Context Protocol: a universal adapter for connecting LLMs to external systems. Instead of building N×M custom integrations (N applications × M data sources), you build N+M: each application implements the MCP client once, each tool implements the server once, and everything interoperates.\n\nWithin a year, MCP achieved something rare: cross-competitor adoption. OpenAI, Google, and Microsoft all support it. SDKs exist for Python, TypeScript, Go, Rust, and more. The community has built thousands of servers covering everything from GitHub to Salesforce to local filesystems.\n\n## The MCP Protocol\n\nMCP servers allow LLMs to discover and use tools at runtime. For every tool, you get a description and information about the input- and output-formats. The LLM uses this information to select the appropriate tool for the current task.\n\nThis allows you to dynamically implement tools and provide them to the LLM. But while MCP solves the problem of tool discovery, it introduces a new one: context composition.\n\nHere's what happens when you invoke an MCP-enabled agent: the context window gets composed of three parts: The user prompt, your system instructions, and all connected MCP tool definitions.\n\nAt Hyground, we connected our AI Ops agents to the tools they needed: log and metric analysis, documentation integration, infrastructure provider integration, and more.\n\n## The 87% context that is not your prompt\n\n87% of our context was MCP tool definitions. 11.4% was instructions. The user's actual prompt? 1.6%.\n\nThis isn't a Hyground-specific problem. The MCP specification requires all tool definitions to be loaded upfront. There's no native mechanism for semantic filtering or lazy loading. Every connected server dumps its full schema into context before the LLM sees a single user token.\n\nThe consequences compound. Major clients have imposed hard limits: Cursor caps at 40 tools, GitHub Copilot at 128. These caps exist because LLM performance degrades when selecting from large, flat tool lists. The model wastes attention on irrelevant tool descriptions, and intermediate tool results further bloat the context.\n\n## The Solution: Dynamic tool discovery\n\nThe industry is converging on a pattern: don't load all tools upfront. Instead, give the agent a discovery mechanism.",[191,192],"AI","Prompt","2026-02-25","2026-05-04T13:39:03.272Z","2026-05-15T14:16:53.342Z","2026-05-15T14:16:53.405Z",{"id":198,"documentId":199,"name":200,"alternativeText":201,"caption":12,"focalPoint":12,"width":202,"height":203,"formats":204,"hash":227,"ext":47,"mime":50,"size":228,"url":229,"previewUrl":12,"provider":28,"provider_metadata":12,"createdAt":230,"updatedAt":231,"publishedAt":230},202,"q7axf0cnw9hwf9a3hnasjn2z","69be7021499e1af91869dc61_Hyground-Thumbnail-stock.webp","Hyground Blogpost Thumbnail",788,526,{"small":205,"medium":212,"thumbnail":219},{"ext":47,"url":206,"hash":207,"mime":50,"name":208,"path":12,"size":209,"width":61,"height":210,"sizeInBytes":211},"\u002Fuploads\u002Fsmall_69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe.webp","small_69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe","small_69be7021499e1af91869dc61_Hyground-Thumbnail-stock.webp",1.19,334,1194,{"ext":47,"url":213,"hash":214,"mime":50,"name":215,"path":12,"size":216,"width":69,"height":217,"sizeInBytes":218},"\u002Fuploads\u002Fmedium_69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe.webp","medium_69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe","medium_69be7021499e1af91869dc61_Hyground-Thumbnail-stock.webp",1.96,501,1962,{"ext":47,"url":220,"hash":221,"mime":50,"name":222,"path":12,"size":223,"width":224,"height":225,"sizeInBytes":226},"\u002Fuploads\u002Fthumbnail_69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe.webp","thumbnail_69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe","thumbnail_69be7021499e1af91869dc61_Hyground-Thumbnail-stock.webp",0.55,234,156,552,"69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe",2.27,"\u002Fuploads\u002F69be7021499e1af91869dc61_Hyground_Thumbnail_stock_85b0d6ecfe.webp","2026-05-15T14:08:48.968Z","2026-05-15T14:14:43.706Z",[],{"pagination":234},{"page":113,"pageSize":235,"pageCount":113,"total":113},25,"\u003Cp>When OpenAI introduced function calling in June 2023, it felt like the missing piece for building useful AI agents. Finally, LLMs could interact with the real world. But anyone who shipped production systems quickly learned the truth: it was finicky. You had to manage the tool call loop yourself, handle errors gracefully, and hope the model picked the right function from your carefully crafted definitions.\u003C\u002Fp>\n\u003Cp>Then came MCP.\u003C\u002Fp>\n\u003Cp>In November 2024, Anthropic open-sourced the Model Context Protocol: a universal adapter for connecting LLMs to external systems. Instead of building N×M custom integrations (N applications × M data sources), you build N+M: each application implements the MCP client once, each tool implements the server once, and everything interoperates.\u003C\u002Fp>\n\u003Cp>Within a year, MCP achieved something rare: cross-competitor adoption. OpenAI, Google, and Microsoft all support it. SDKs exist for Python, TypeScript, Go, Rust, and more. The community has built thousands of servers covering everything from GitHub to Salesforce to local filesystems.\u003C\u002Fp>\n\u003Ch2>The MCP Protocol\u003C\u002Fh2>\n\u003Cp>MCP servers allow LLMs to discover and use tools at runtime. For every tool, you get a description and information about the input- and output-formats. The LLM uses this information to select the appropriate tool for the current task.\u003C\u002Fp>\n\u003Cp>This allows you to dynamically implement tools and provide them to the LLM. But while MCP solves the problem of tool discovery, it introduces a new one: context composition.\u003C\u002Fp>\n\u003Cp>Here's what happens when you invoke an MCP-enabled agent: the context window gets composed of three parts: The user prompt, your system instructions, and all connected MCP tool definitions.\u003C\u002Fp>\n\u003Cp>At Hyground, we connected our AI Ops agents to the tools they needed: log and metric analysis, documentation integration, infrastructure provider integration, and more.\u003C\u002Fp>\n\u003Ch2>The 87% context that is not your prompt\u003C\u002Fh2>\n\u003Cp>87% of our context was MCP tool definitions. 11.4% was instructions. The user's actual prompt? 1.6%.\u003C\u002Fp>\n\u003Cp>This isn't a Hyground-specific problem. The MCP specification requires all tool definitions to be loaded upfront. There's no native mechanism for semantic filtering or lazy loading. Every connected server dumps its full schema into context before the LLM sees a single user token.\u003C\u002Fp>\n\u003Cp>The consequences compound. Major clients have imposed hard limits: Cursor caps at 40 tools, GitHub Copilot at 128. These caps exist because LLM performance degrades when selecting from large, flat tool lists. The model wastes attention on irrelevant tool descriptions, and intermediate tool results further bloat the context.\u003C\u002Fp>\n\u003Ch2>The Solution: Dynamic tool discovery\u003C\u002Fh2>\n\u003Cp>The industry is converging on a pattern: don't load all tools upfront. Instead, give the agent a discovery mechanism.\u003C\u002Fp>\n"]