Quickstart

Answeringhttps://mttr.si/mcpStreamable HTTPJSON API at /api/v1

1. 1
   Get a key

   Free, with an email. It reads any site's public report; claim your own site to add its private parts.

   Your emailGet a key

   Or connect with no key, and your agent signs you up itself with mttr_signup.
2. 2
   Connect where your agent runs

   First put the key in your shell, where no file you commit can hold it.

   export MTTR_KEY=mk_your_key

   Run this once, from any folder. Claude Code keeps it for every project.

   claude mcp add --transport http mttr https://mttr.si/mcp \
     --header "Authorization: Bearer $MTTR_KEY"

   First put the key in your shell, where no file you commit can hold it.

   export MTTR_KEY=mk_your_key

   Add this to Codex’s settings. It reads the key from your shell when it starts.

   [mcp_servers.mttr]
   url = "https://mttr.si/mcp"
   bearer_token_env_var = "MTTR_KEY"

   Add this to Gemini CLI’s settings, then start it again and run /mcp to see mttr’s tools.

   {
     "mcpServers": {
       "mttr": {
         "httpUrl": "https://mttr.si/mcp",
         "headers": { "Authorization": "Bearer mk_your_key" }
       }
     }
   }

   The key sits in the settings file above. Keep that file to yourself.

   "Authorization": "Bearer mk_your_key"

   Cursor reads the key from the shell it was opened from.

   export MTTR_KEY=mk_your_key

   Add this in the site’s repository, then turn mttr on under MCP in Cursor’s settings.

   {
     "mcpServers": {
       "mttr": {
         "url": "https://mttr.si/mcp",
         "headers": { "Authorization": "Bearer ${env:MTTR_KEY}" }
       }
     }
   }

   Add this under Settings, Customizations, MCP servers, or in this file; then refresh the list of servers. In one project only, use .agents/mcp_config.json instead.

   {
     "mcpServers": {
       "mttr": {
         "serverUrl": "https://mttr.si/mcp",
         "headers": { "Authorization": "Bearer mk_your_key" }
       }
     }
   }

   The key sits in the settings file above. Keep that file to yourself, and out of any repository.

   "Authorization": "Bearer mk_your_key"

   Add this in the site’s repository. VS Code asks for the key the first time and keeps it safe.

   {
     "servers": {
       "mttr": {
         "type": "http",
         "url": "https://mttr.si/mcp",
         "headers": { "Authorization": "Bearer ${input:mttr-key}" }
       }
     },
     "inputs": [
       { "id": "mttr-key", "type": "promptString", "description": "Your mttr key", "password": true }
     ]
   }

   Nothing to set by hand: paste the key when VS Code asks for it.

   Your mttr key: mk_your_key

   Add this to Windsurf’s MCP settings, then refresh its list of servers.

   {
     "mcpServers": {
       "mttr": {
         "serverUrl": "https://mttr.si/mcp",
         "headers": { "Authorization": "Bearer mk_your_key" }
       }
     }
   }

   The key sits in the settings file above. Keep that file to yourself.

   "Authorization": "Bearer mk_your_key"

   Any client that speaks MCP over HTTP connects with these three things.

   Address    https://mttr.si/mcp
   Transport  Streamable HTTP
   Header     Authorization: Bearer mk_your_key

   Every request carries the key as a bearer token.

   Authorization: Bearer mk_your_key

3. 3
   Ask it something

   In words. Your agent picks the call, and the answer is the one a person sees on the report.

   • What should I fix first on example.com?mttr_brief({ site: "example.com" })
   • Fix the top issue, then have mttr read the site again.mttr_rescan({ site: "example.com" })
   • Who am I up against, and what do they cover that I don’t?mttr_competitors({ site: "example.com" })

   What comes backThe start of a brief, a sample

   What mttr_brief answersThe start of one, a sample

   # Fix list for example.com
   
   mttr read 64 pages of https://example.com on 2026-10-15 and scored them 72 of 100.
   
   A score says pages are well formed. It does not say an assistant cites the site.
   
   ## 1. Put the text in the HTML
   
   - **Failing:** 9 pages
   - **What to do:** Put the page text in the HTML the server sends. Most AI crawlers do not run scripts.
   - **Pages:**
     - `/services/drainage`
     - `/pricing`
     - and 7 more