What is MCP?
The Model Context Protocol (MCP) is an open standard for connecting AI assistants to external tools and data. An MCP server exposes a set of tools that any compatible client — Claude, Cursor, n8n, custom agents — can discover and call.
The Omnflow MCP server exposes our content repurposing and Voice DNA engines so you can run them without leaving your AI tool of choice. Learn more at modelcontextprotocol.io.
Connection details
Server URL
Server name
Transport
Streamable HTTP (POST)
Auth
OAuth 2.1 (PKCE) — sign in with your Omnflow account on connect
Available tools
repurpose_content
Turn a source idea, blog post, or transcript into 5 platform-ready posts (Twitter/X thread, LinkedIn post, Instagram caption, email newsletter, TikTok script).
Parameters
source_text — string (40–8000 chars)
analyze_voice
Analyze a writing sample and return the author's Voice DNA: tone, sentence style, what they avoid, and signature move.
Parameters
sample — string (100–2000 chars)
list_recent_demos
List the most recent public demo runs (source text + generated outputs + inferred niche).
Parameters
limit — integer (1–20, default 5)
save_to_library
Save a piece of repurposed content to your Omnflow content library.
Parameters
content, platform (required); title, source_text, tags (optional)
add_to_calendar
Schedule a post on a specific date in your Omnflow content calendar.
Parameters
content, platform, scheduled_date (required); scheduled_time, title (optional)
list_voice_connections
List your connected learning sources (ChatGPT, Claude), past import sessions, and current Writing DNA / Thinking DNA strength.
Parameters
none
start_voice_import
Begin an import session that teaches Omnflow from your own writing inside a chat tool. Returns a session_id plus the rules for what may be sent.
Parameters
source — "chatgpt" | "claude"
add_voice_import_items
Add conversations to the session. Send only your own messages, concatenated per conversation — never assistant replies. Duplicates are skipped.
Parameters
session_id, items[] (title, content, date)
preview_voice_import
Show what Omnflow would learn from the session (word counts, Writing vs Thinking DNA buckets) before anything is committed.
Parameters
session_id
commit_voice_import
Merge the session into your Voice DNA and return which dimensions were strengthened. Always confirm with the user first.
Parameters
session_id
delete_voice_import
Delete an import session's material and roll back its contribution to your Voice DNA.
Parameters
session_id
Install
Pick your client. The same server URL works everywhere.
Edit claude_desktop_config.json:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
claude_desktop_config.json
{
"mcpServers": {
"omnflow": {
"url": "https://omnflow.com/api/mcp"
}
}
}Restart Claude Desktop. The three Omnflow tools appear in the tools menu.
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..."Analyze your writing voice
Use omnflow's analyze_voice tool on this sample:
"[paste 100–2000 characters of your writing]"Inspect recent demo runs
Use omnflow's list_recent_demos to show the last 10 demo runs.Use cases
Solopreneurs
Turn one blog post into a week of social content without leaving Claude.
Coaches
Repurpose podcast or video transcripts into LinkedIn and Twitter threads in seconds.
Agencies
Standardize client voice with analyze_voice, then automate output through n8n.
Content teams
Wire Omnflow into Cursor or custom agents to draft, refine, and ship faster.
FAQ
Ready to plug Omnflow into your AI stack?
Free to use. No API key. Three tools, infinite content.