OmniSocialsOmniSocials
DocumentationAPI reference
Help centerSupport
  • Documentation
  • API reference
Developers
  • Quickstart
  • API reference
  • SDKs
  • MCP server
  • CLI
For AI agents
  • Docs for AI agents
  • llms.txt
  • OpenAPI spec
  • Postman collection
Product
  • OmniSocials
  • Pricing
  • Changelog
  • Help center
Company
  • Support
  • Privacy policy
  • Terms of service

© 2026 OmniSocials

OmniSocials on GitHubOmniSocials on XOmniSocials on LinkedIn
Overview
Get started
SDKs
CLI
MCPDocs for AI agentsWebhooks
Guides
Platforms
Integrations
Migrate
Plans and billing
GlossaryChangelogAPI reference
Documentation

Use the OmniSocials docs from an AI agent

Every page of these docs is also available in a form an agent can read without the website. Pick the one your tool supports.

WhatURLUse it for
Indexllms.txtEvery page with a one-line summary, grouped like the sidebar
Full textllms-full.txtAll guide pages in one file
Markdown pagesAdd .md to a page URL, like /introduction.mdOne page, without the site around it
Content negotiationSend Accept: text/markdown to a page URLThe same markdown at the normal URL
Docs MCP serverhttps://docs.omnisocials.com/mcpSearch and read the docs from an MCP client
OpenAPI specopenapi.jsonEvery REST endpoint, for code generation and API tools
Postman collectionomnisocials.postman_collection.jsonImport into Postman, Insomnia or Bruno
Agent skillskill.mdShort working instructions for a coding agent

Two MCP servers

OmniSocials runs two MCP servers. They do different jobs.

  • The docs MCP server (https://docs.omnisocials.com/mcp) reads these docs. It is read-only and needs no sign-in.
  • The OmniSocials MCP server (https://mcp.omnisocials.com) acts on your workspace: it creates and schedules posts, reads analytics and answers the inbox. It needs a sign-in or an API key. See MCP server.

A coding agent that builds an integration usually wants both: the docs server to look things up, and the API itself (or the OmniSocials MCP server) to test against.

Connect the docs MCP server

The server speaks MCP over Streamable HTTP and has four tools:

ToolWhat it does
search_docsSearches the guides, platform pages, SDK pages and the API reference
get_pageReturns one page as markdown, by path or URL
get_endpointReturns the full OpenAPI definition of one endpoint, with schemas resolved
list_pagesReturns the index (llms.txt)

Claude Code

TerminalTerminal
claude mcp add --transport http omnisocials-docs https://docs.omnisocials.com/mcp

Cursor

Open this install link in Cursor, or add the server to ~/.cursor/mcp.json:

JSON
{ "mcpServers": { "omnisocials-docs": { "url": "https://docs.omnisocials.com/mcp" } } }

VS Code

Add the server to .vscode/mcp.json in your project:

JSON
{ "servers": { "omnisocials-docs": { "type": "http", "url": "https://docs.omnisocials.com/mcp" } } }

Other clients

Any MCP client that supports Streamable HTTP can connect to https://docs.omnisocials.com/mcp. No headers or keys are needed.

Fetch a page as markdown

Both requests return the same markdown:

TerminalTerminal
curl https://docs.omnisocials.com/creating-posts.md curl -H "Accept: text/markdown" https://docs.omnisocials.com/creating-posts

The home page URL with Accept: text/markdown returns the llms.txt index.

From a page in the browser

Every guide page has a Copy page button. Its menu copies a link, opens the markdown version, or opens the page in Claude or ChatGPT with the page as context.

Last modified on October 21, 2026
MCPWebhooks
On this page
  • Two MCP servers
  • Connect the docs MCP server
    • Claude Code
    • Cursor
    • VS Code
    • Other clients
  • Fetch a page as markdown
  • From a page in the browser
JSON
JSON