{"Status":true,"Message":"","Response":{"post":{"postuid":"ac0449df-1423-4ab8-9e09-e886f5ba34e5","tenantuid":"d8b744fc-2e70-4089-bb80-dd1d08f6c7b2","projectuid":"fc6490ac-7527-4f49-b06e-46f701280e85","title":"Model Context Protocol","slug":"article/model-context-protocol","html":"\u003Ccite class=\u0022important\u0022\u003E\u003Cspan class=\u0022title\u0022\u003EImportant\u003C/span\u003E\u003Cp\u003EYou may have been redirected to this page when attempting to access the /mcp path to your HelpGuides.io documentation.\u003C/p\u003E\u003C/cite\u003E\u003Ch2 id=\u0022what_is_model_context_protocol\u0022\u003EWhat is Model Context Protocol\u003C/h2\u003E\u003Cp\u003EModel Context Protocol (MCP) is an emerging open standard that defines how AI systems interact with external tools and knowledge sources in a structured, predictable, and programmatic way. \u003C/p\u003E\u003Cp\u003ERather than relying on scraping web pages or processing unstructured text, MCP allows AI clients to discover available capabilities, query systems using well-defined inputs, and retrieve results in consistent, machine-readable formats. \u003C/p\u003E\u003Cp\u003EThis enables AI assistants, agents, and retrieval-augmented generation (RAG) workflows to integrate external context reliably, turning documentation and services into first-class components of an enterprise AI stack while reducing fragility, ambiguity, and long-term integration risk.\u003C/p\u003E\u003Ch2 id=\u0022mcp_support_simplifies_integration_for_ragbased_scenarios_\u0022\u003EMCP Support Simplifies Integration for RAG-based scenarios \u003C/h2\u003E\u003Cp\u003EContent published in \u003Ca target=\u0022_self\u0022 href=\u0022http://helpguides.io/\u0022\u003EHelpGuides.io\u003C/a\u003E is automatically structured and chunked at publish time. Long articles are broken into logically related sections based on headings, then tokenized and indexed as discrete units. \u003C/p\u003E\u003Cp\u003EThis approach allows search to retrieve only the most relevant portions of content for a given query, improving response accuracy.\u003C/p\u003E\u003Cp\u003EYou can see this when searching for content in the UX as search can show relative sections. For example, there are multiple results returned for the article Setting up Text Message Marketing:\u003C/p\u003E\u003Cimg class=\u0022lazy-load\u0022 data-width=\u00221161\u0022 data-height=\u00221000\u0022 data-src=\u0022https://graffiti-auf7e6dwhxhcbwek.z03.azurefd.net/d8b744fc-2e70-4089-bb80-dd1d08f6c7b2/fc6490ac-7527-4f49-b06e-46f701280e85/f8a3d494-7c31-4101-98c9-7f21d41e1735.png?v=1986912245\u0022 style=\u0022\u0022 alt=\u0022Search results showing several sections of the same article, each with its own heading and excerpt\u0022/\u003E\u003Cp\u003EWith support for Model Context Protocol (MCP), downstream AI tools, such as Microsoft Copilot when integrated via MCP, benefit from this same structure content approach. And most importantly, no longer need to rely on scraping HTML. \u003C/p\u003E\u003Cp\u003EInstead, documentation is exposed through a clean, structured, and pre-chunked interface, ensuring that only the content most relevant to a given query is returned. In practice, MCP enables applications to interact with structured knowledge sources more intelligently, retrieving precise context optimized for retrieval-augmented generation (RAG).\u003C/p\u003E\u003Ch2 id=\u0022turning_on_mcp_and_finding_your_url\u0022\u003ETurning on MCP and finding your URL\u003C/h2\u003E\u003Cp\u003EOpen your project, click \u003Cb\u003EAI Settings\u003C/b\u003E and stay on the \u003Cb\u003EModel Context Protocol\u003C/b\u003E tab. Make sure \u003Cb\u003EEnable Model Context Protocol\u003C/b\u003E is on, then copy \u003Cb\u003EModel Context Protocol URL for this project\u003C/b\u003E. It\u0027s your site\u0027s domain followed by \u003Ccode\u003E/mcp\u003C/code\u003E:\u003C/p\u003E\u003Cdiv class=\u0022code_wrapper\u0022\u003E\u003Cdiv class=\u0022code\u0022 data-language=\u0022text\u0022\u003Ehttps://YOUR-DOMAIN/mcp\u003C/div\u003E\u003C/div\u003E\u003Cp\u003EAI assistants sign in with OAuth: the first time you connect one, you sign in to HelpGuides and approve its access. When the switch is off, the URL stops responding and connected assistants lose access. See \u003Ca href=\u0022/article/ai-settings\u0022\u003EAI Settings\u003C/a\u003E and, for every tool the endpoint offers, \u003Ca href=\u0022/model-context-protocol-endpoints\u0022\u003EModel Context Protocol Endpoints\u003C/a\u003E.\u003C/p\u003E\u003Cfigure\u003E\u003Cimg class=\u0022lazy-load\u0022 data-src=\u0022https://graffiti-auf7e6dwhxhcbwek.z03.azurefd.net/d8b744fc-2e70-4089-bb80-dd1d08f6c7b2/fc6490ac-7527-4f49-b06e-46f701280e85/images/335d2d8b5780455ba153aa995f951576.png?v=676174577\u0022 data-width=\u00221150\u0022 data-height=\u0022640\u0022 style=\u0022max-width: 100%;\u0022 alt=\u0022AI Settings Model Context Protocol tab with the MCP URL and the Enable switch\u0022/\u003E\u003Cfigcaption\u003EAI Settings, Model Context Protocol tab\u003C/figcaption\u003E\u003C/figure\u003E\u003Ch2 id=\u0022connecting_your_mcp_to_llms\u0022\u003EConnecting your MCP to LLMs\u003C/h2\u003E\u003Cp\u003EYou can connect your MCP URL to any AI assistant that supports remote MCP servers. See the instructions for:\u003C/p\u003E\u003Cul\u003E\u003Cli\u003E\u003Ca href=\u0022/article/adding-helpguides-to-claude-as-an-mcp-connector\u0022\u003EClaude MCP setup instructions\u003C/a\u003E\u003C/li\u003E\u003Cli\u003E\u003Ca href=\u0022/article/add-chatgpt-as-an-mcp-connector\u0022\u003EChatGPT MCP setup instructions\u003C/a\u003E\u003C/li\u003E\u003C/ul\u003E\u003Ch2 id=\u0022testing_mcp_via_curl\u0022\u003ETesting MCP via CURL\u003C/h2\u003E\u003Cp\u003EYou can test your MCP endpoint with curl. Every request needs an OAuth access token in an \u003Ccode\u003EAuthorization: Bearer\u003C/code\u003E header. Without one, the endpoint replies \u003Ccode\u003E401\u003C/code\u003E with a \u003Ccode\u003EWWW-Authenticate\u003C/code\u003E header that tells MCP clients where to sign in.\u003C/p\u003E\u003Cp\u003ENote, in practice, these interactions are handled automatically: an MCP client queries the server, discovers the supported capabilities and endpoints, and then invokes the appropriate endpoint to fulfill a request.\u003C/p\u003E\u003Ch3 id=\u0022initial_request\u0022\u003EInitial request\u003C/h3\u003E\u003Cp\u003EThe first request initializes a session. It returns the server info and an \u003Cb\u003EMcp-Session-Id\u003C/b\u003E header, which you send with every later request:\u003C/p\u003E\u003Cdiv class=\u0022code_wrapper\u0022\u003E\u003Cdiv class=\u0022code\u0022 data-language=\u0022sh\u0022\u003Ecurl -i -X POST https://YOUR-DOMAIN/mcp \\\n  -H \u0026quot;Authorization: Bearer ACCESS_TOKEN\u0026quot; \\\n  -H \u0026quot;Content-Type: application/json\u0026quot; \\\n  -H \u0026quot;Accept: application/json, text/event-stream\u0026quot; \\\n  -d \u0026#39;{\u0026quot;jsonrpc\u0026quot;:\u0026quot;2.0\u0026quot;,\u0026quot;id\u0026quot;:1,\u0026quot;method\u0026quot;:\u0026quot;initialize\u0026quot;,\u0026quot;params\u0026quot;:{\u0026quot;protocolVersion\u0026quot;:\u0026quot;2024-11-05\u0026quot;,\u0026quot;capabilities\u0026quot;:{},\u0026quot;clientInfo\u0026quot;:{\u0026quot;name\u0026quot;:\u0026quot;Example\u0026quot;,\u0026quot;version\u0026quot;:\u0026quot;1.0.0\u0026quot;}}}\u0026#39;\u003C/div\u003E\u003C/div\u003E\u003Cp\u003EThe response includes the \u003Cb\u003EMcp-Session-Id\u003C/b\u003E header:\u003C/p\u003E\u003Cdiv class=\u0022code_wrapper\u0022\u003E\u003Cdiv class=\u0022code\u0022 data-language=\u0022sh\u0022\u003EHTTP/1.1 200 OK\r\nContent-Type: text/event-stream\r\nDate: Fri, 16 Jan 2026 17:25:08 GMT\r\nServer: Kestrel\r\nCache-Control: no-cache,no-store\r\nContent-Encoding: identity\r\nTransfer-Encoding: chunked\r\nContent-Security-Policy: frame-ancestors \u0026#39;none\u0026#39;\r\nX-Frame-Options: DENY\r\nMcp-Session-Id: MM2ZUNNpVih7r0TGZb38pg\r\n\r\nevent: message\r\ndata: {\u0026quot;result\u0026quot;:{\u0026quot;protocolVersion\u0026quot;:\u0026quot;2024-11-05\u0026quot;,\u0026quot;capabilities\u0026quot;:{\u0026quot;logging\u0026quot;:{},\u0026quot;tools\u0026quot;:{\u0026quot;listChanged\u0026quot;:true}},\u0026quot;serverInfo\u0026quot;:{\u0026quot;name\u0026quot;:\u0026quot;io.helpguides/sever\u0026quot;,\u0026quot;version\u0026quot;:\u0026quot;1.0.0\u0026quot;}},\u0026quot;id\u0026quot;:1,\u0026quot;jsonrpc\u0026quot;:\u0026quot;2.0\u0026quot;}\u003C/div\u003E\u003C/div\u003E\u003Ch3 id=\u0022perform_a_query\u0022\u003EPerform a query\u003C/h3\u003E\u003Cp\u003ENext, call a tool using the session ID. This example calls the \u003Ccode\u003Esearch\u003C/code\u003E tool:\u003C/p\u003E\u003Cdiv class=\u0022code_wrapper\u0022\u003E\u003Cdiv class=\u0022code\u0022 data-language=\u0022sh\u0022\u003Ecurl -X POST https://YOUR-DOMAIN/mcp \\\n  -H \u0026quot;Authorization: Bearer ACCESS_TOKEN\u0026quot; \\\n  -H \u0026quot;Content-Type: application/json\u0026quot; \\\n  -H \u0026quot;Accept: application/json, text/event-stream\u0026quot; \\\n  -H \u0026quot;Mcp-Session-Id: YOUR_MCP_SESSION_ID\u0026quot; \\\n  -d \u0026#39;{\u0026quot;jsonrpc\u0026quot;:\u0026quot;2.0\u0026quot;,\u0026quot;id\u0026quot;:2,\u0026quot;method\u0026quot;:\u0026quot;tools/call\u0026quot;,\u0026quot;params\u0026quot;:{\u0026quot;name\u0026quot;:\u0026quot;search\u0026quot;,\u0026quot;arguments\u0026quot;:{\u0026quot;s\u0026quot;:\u0026quot;setting up SMS marketing\u0026quot;}}}\u0026#39;\u003C/div\u003E\u003C/div\u003E\u003Cp\u003EFor the full list of tools and their parameters, see \u003Ca href=\u0022/model-context-protocol-endpoints\u0022\u003EModel Context Protocol Endpoints\u003C/a\u003E. To have an agent act on instructions you leave in the editor, see \u003Ca href=\u0022/article/review-comments-and-ai-agents\u0022\u003EReview Comments and AI Agents\u003C/a\u003E.\u003C/p\u003E","publish_status":0,"post_type":"Article","authoruid":"3dde8c16-763a-4a2b-ae0b-1d8c50c62e3d","author":{"authoruid":"3dde8c16-763a-4a2b-ae0b-1d8c50c62e3d"},"featured_image_updating":false,"meta_description":"Model Context Protocol (MCP) lets AI access structured HelpGuides.io docs without scraping, enabling smarter, reliable retrieval for RAG workflows and AI tools.","display_toc":true,"has_workingcopy":false,"allow_indexing":true,"total_views":5845,"date_published":"2026-01-16T17:21:00","date_updated":"2026-10-04T21:48:21.177","date_created":"2026-01-16T17:18:07.323"}}}