mcpField
O
OMNFLOW

OMNFLOW

github.com/ericharrisonecho-limaOther0Updated 18w ago
Repository

About OMNFLOW

OMNFLOW is a content-repurposing service for creators, marketers, and founders. Its MCP server turns source material into platform-specific posts and can save or schedule them through an MCP client.

Key features

  1. 01Repurpose text into five post formats
  2. 02Analyze a writer’s voice
  3. 03Save posts to a content library
  4. 04Schedule posts on a calendar
  5. 05Review recent demo runs

Use cases

  1. 01Adapt a blog post for social platforms
  2. 02Generate content from a transcript or idea
  3. 03Save generated posts for later use
  4. 04Schedule posts in a content calendar

MCP support

The remote MCP server provides tools for repurposing content, analyzing voice, viewing demos, saving posts, and scheduling them. MCP-compatible clients connect over HTTP and authenticate with OAuth 2.1.

Details

From the project README

Omnflow MCP Server

Turn any content into platform-ready posts for Twitter/X, LinkedIn, Instagram, Email, and TikTok — save them to your library, and schedule them on your content calendar. Powered by Omnflow.

MCP License: MIT

What is Omnflow?

Omnflow is an AI content repurposing tool for solopreneurs, coaches, and creators. Paste one blog post, video transcript, podcast, or idea and Omnflow rewrites it for X/Twitter, LinkedIn, Instagram, TikTok, YouTube Shorts, and email — in your voice.

The Omnflow MCP server brings these capabilities directly into Claude, Cursor, n8n, and any other MCP-compatible client — including saving generated posts to your Omnflow library and scheduling them on your content calendar.

Connection

  • Server URL: https://omnflow.com/api/mcp
  • Transport: Streamable HTTP (POST)
  • Auth: OAuth 2.1 (PKCE) — sign in with your Omnflow account on connect
  • Name: omnflow

Tools

ToolDescriptionParameters
repurpose_contentTurn a source idea, blog post, or transcript into 5 platform-ready posts (Twitter thread, LinkedIn post, Instagram caption, email newsletter, TikTok script).source_text (string, 40–8000 chars)
analyze_voiceAnalyze a writing sample and return the author's Voice DNA: tone, sentence style, what they avoid, and their signature move.sample (string, 100–2000 chars)
list_recent_demosList the most recent public demo runs (source text + generated outputs + inferred niche).limit (integer, 1–20, default 5)
save_to_librarySave a piece of content (post, caption, thread, script) directly into your Omnflow library so it shows up on the Library page.content (string), platform (twitter | linkedin | instagram | tiktok | email), title? (string), tags? (string[])
add_to_calendarSchedule a post on your Omnflow content calendar at a specific time.content (string), platform (same enum), scheduled_at (ISO 8601 datetime), title? (string)

Installation

Claude Desktop

Edit claude_desktop_config.json:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "omnflow": {
      "url": "https://omnflow.com/api/mcp"
    }
  }
}

Restart Claude Desktop. On first tool call, a browser window opens for you to sign in to Omnflow — after that the five tools are available in the tools menu.

Cursor

In Cursor Settings → MCP → Add new MCP server:

{
  "mcpServers": {
    "omnflow": {
      "url": "https://omnflow.com/api/mcp"
    }
  }
}

Cursor will trigger the same OAuth sign-in flow on first use.

Claude Code CLI
claude mcp add --transport http omnflow https://omnflow.com/api/mcp
n8n

Use the MCP Client node (or HTTP Request node):

  • Endpoint: https://omnflow.com/api/mcp
  • Method: POST
  • Headers:
    • Content-Type: application/json
    • Accept: application/json, text/event-stream
    • Authorization: Bearer <access_token> (obtain via the OAuth flow described below)
Claude.ai (web) — Custom Connector

In Claude.ai → Settings → Connectors → Add custom connector, paste:

https://omnflow.com/api/mcp

Claude discovers OAuth via /.well-known/oauth-protected-resource, opens a browser window to https://omnflow.com/auth, asks you to sign in or create an Omnflow account, and then stores the access token. After that all five tools are available in any chat.

Authentication

This server uses the OAuth 2.1 Authorization Code flow with PKCE. You don't need an API key — when you connect Omnflow, your MCP client opens a browser window so you can sign in to your Omnflow account. All five tools require authentication. Unauthenticated calls return 401 Unauthorized with a WWW-Authenticate header pointing at the discovery document so compliant clients can re-trigger the sign-in flow.

Discovery endpoints:

  • https://omnflow.com/.well-known/oauth-authorization-server
  • https://omnflow.com/.well-known/oauth-protected-resource

Endpoints used by the flow:

  • Authorization: https://omnflow.com/auth
  • Token: https://omnflow.com/api/auth/token
  • Scopes: read, write (granted together — both are required for save / schedule tools)

Usage Examples

Repurpose a blog post
Use omnflow's repurpose_content tool on this blog post:

"The biggest mistake I see founders make is shipping features
nobody asked for. Here's the 3-step process I use instead..."

Returns: a Twitter thread, LinkedIn post, Instagram caption, email newsletter, and TikTok script.

Save a generated post to your library
Use omnflow's save_to_library to save this LinkedIn post to my library
with the tag "founder-lessons":

"The biggest mistake I see founders make is..."

The post appears on your Omnflow Library page immediately.

Schedule a post on the calendar
Use omnflow's add_to_calendar to schedule this tweet for
2026-07-01T09:00:00Z:

"Ship the boring thing. The exciting thing can wait."

The post shows up on your Omnflow Calendar at the scheduled time.

Analyze writing voice
Use omnflow's analyze_voice tool on this sample:

"[paste 100-2000 characters of your writing]"

Returns Voice DNA: tone, sentence_style, avoids, signature_move.

Inspect recent demo runs
Use omnflow's list_recent_demos to show the last 10 demo runs.

Links

License

MIT In Cursor Settings → MCP → Add new MCP server:

{
  "mcpServers": {
    "omnflow": {
      "url": "https://omnflow.com/api/mcp"
    }
  }
}
Claude Code CLI
claude mcp add --transport http omnflow https://omnflow.com/api/mcp
n8n

Use the MCP Client node (or HTTP Request node):

  • Endpoint: https://omnflow.com/api/mcp
  • Method: POST
  • Headers:
    • Content-Type: application/json
    • Accept: application/json, text/event-stream
Claude.ai (web)

The web "Custom Connectors" UI requires OAuth discovery, which this server does not yet expose. Use Claude Desktop instead.

Usage Examples

Repurpose a blog post
Use omnflow's repurpose_content tool on this blog post:

"The biggest mistake I see founders make is shipping features
nobody asked for. Here's the 3-step process I use instead..."

Returns: a Twitter thread, LinkedIn post, Instagram caption, email newsletter, and TikTok script.

Analyze writing voice
Use omnflow's analyze_voice tool on this sample:

"[paste 100-2000 characters of your writing]"

Returns Voice DNA: tone, sentence_style, avoids, signature_move.

Inspect recent demo runs
Use omnflow's list_recent_demos to show the last 10 demo runs.

Links

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.