RightBlogger Developer Resources

RightBlogger API: the MCP server

RightBlogger's programmatic interface is a remote Model Context Protocol (MCP) server. Connect it to Claude, ChatGPT, Cursor, or any MCP client to query your Google Search Console performance, research keywords from RightBlogger's keyword database, check your AI brand visibility, read and edit your CMS posts, and manage your content calendar, automations, social posts, and Site Agent. Every tool is scoped to the projects your account can access, anything that publishes or uses your plan asks you to confirm first, and it is included on all plans.

https://rightblogger.com/api/mcp

Authentication is OAuth 2.1 with PKCE and dynamic client registration. There are no API keys: every credential is bound to a RightBlogger account, and users approve access with a one-time email code. Agents can follow auth.md end to end without a human in the loop until the sign-in step.

Full MCP connector docs, setup, and tool reference

Access and onboarding

Fully self-serve, for humans and agents alike. There is a free plan to start on, there are no API keys to provision, and there is no sales call: an agent registers its own OAuth client via dynamic client registration (RFC 7591), sends the user through a one-time email code sign-in, and holds a token scoped to that account. The MCP connector is included on every plan.

Machine-readable endpoints

Stable URLs an agent can fetch directly. All of them are public and need no authentication.

• /get-started.md

  text/markdown

  get-started.md. Agent onboarding: tell any AI agent "Set up RightBlogger for me. Fetch https://rightblogger.com/get-started.md and follow it." and it connects the MCP server and walks the user through setup.

• /openapi.json

  application/json

  OpenAPI spec. OpenAPI 3.1 description of every public endpoint: the MCP server, discovery documents, and machine-readable content routes.

• /llms.txt

  text/plain

  llms.txt. Machine-readable product index: what RightBlogger does, pricing, how agents should use it, and links to the help, blog, glossary and tool indexes.

• /llms-full.txt

  text/plain

  llms-full.txt. The product guide and every help article in one Markdown file.

• /product.md

  text/markdown

  product.md. Product guide: every feature, plan limits, integrations, common questions and terms.

• /auth.md

  text/markdown

  auth.md. Agent authorization manifest: how AI agents register an OAuth client and get user-scoped access to the MCP server.

• /.well-known/oauth-protected-resource

  application/json

  OAuth resource metadata. RFC 9728 discovery document naming the OAuth 2.1 authorization server for the MCP endpoint.

• /.well-known/ai-catalog.json

  application/ai-catalog+json

  AI Catalog (experimental). Public discovery index pointing to the MCP Server Card. Follows the experimental AI Catalog draft.

• /api/mcp/server-card

  application/mcp-server-card+json

  MCP Server Card (experimental). Public server identity and connection details. Follows the experimental Server Card draft; account access still requires sign-in.

• /pricing.md

  text/markdown

  pricing.md. Current plans, prices, and plan limits in markdown.

• /sitemap.xml

  application/xml

  sitemap.xml. Every indexable page. Blog, glossary, and help URLs all have markdown alternates: append .md to the URL.

• /feed.xml

  application/rss+xml

  feed.xml. RSS feed of the RightBlogger blog.

API errors

API routes return structured JSON errors with a stable code, a human-readable message, and a hint for where to look next. A request to a nonexistent API path returns HTTP 404 with:

{
  "error": "Not found",
  "code": "NOT_FOUND",
  "message": "No API route exists at /api/example.",
  "hint": "RightBlogger's public API surface is described at https://rightblogger.com/openapi.json...",
  "openapi": "https://rightblogger.com/openapi.json",
  "docs": "https://rightblogger.com/developers"
}