InfluenceFlow MCP Server

A Model Context Protocol (MCP) server that enables Claude to manage influencer marketing campaigns through natural conversation.

Features

• Creator Discovery: Search and filter creators by niche, followers, engagement rate, location, and pricing
• Campaign Management: Full lifecycle — create, activate, pause, complete, and delete campaigns
• AI-Powered Discovery: Trigger creator discovery, get keyword suggestions, bulk invite creators
• Offer Workflow: Send offers to creators, handle negotiations, and track responses
• Application Management: Review, accept, reject, or counter creator applications
• Collaboration Tracking: Monitor content creation from contract to completion
• Content Review: Approve or request revisions on submitted content
• Milestone Management: Track and release payment milestones
• Talent Pools: Organize creators into reusable talent pools
• Tracking Links: Create and monitor affiliate/tracking links
• Webhooks: Configure real-time event notifications
• Analytics: Performance metrics, ROI analysis, and agent usage tracking
• MCP Resources: Live data feeds for campaigns, analytics, collaborations, and webhook events
• MCP Prompts: Guided workflow templates for common tasks

Installation

pip install influenceflow-mcp

Or use with uvx (no installation required):

uvx influenceflow-mcp

Configuration

1. Get Your API Key

1. Log into InfluenceFlow
2. Go to Settings > API Keys
3. Create a new API key

2. Set Environment Variable

export INFLUENCEFLOW_API_KEY="your_api_key_here"

3. Configure Claude Desktop

Add to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "influenceflow": {
      "command": "uvx",
      "args": ["influenceflow-mcp"],
      "env": {
        "INFLUENCEFLOW_API_KEY": "your_api_key_here"
      }
    }
  }
}

Or if installed via pip:

{
  "mcpServers": {
    "influenceflow": {
      "command": "influenceflow-mcp",
      "env": {
        "INFLUENCEFLOW_API_KEY": "your_api_key_here"
      }
    }
  }
}

Available Tools (56 total)

Creator Tools (3)

Campaign Tools (8)

Offer Tools (4)

Application Tools (5)

Collaboration Tools (8)

Content Tools (4)

Milestone Tools (3)

Talent Pool Tools (5)

Tracking Link Tools (3)

Webhook Tools (4)

Discovery Tools (5)

Analytics Tools (4)

MCP Resources

Live read-only data feeds that provide context to Claude:

MCP Prompts

Guided workflow templates for common multi-step tasks:

Agent Identification

If you're building an AI agent that uses the InfluenceFlow API, we recommend setting the X-Agent-Id header to identify your agent:

headers = {
    "Authorization": "Bearer your_api_key",
    "X-Agent-Id": "my-agent/1.0"
}

This enables:

• Agent-specific analytics in the InfluenceFlow dashboard
• Better debugging and support
• Usage tracking via DAA (Daily Active Agents) and MAA (Monthly Active Agents) metrics

The get_agent_usage tool returns these metrics for your brand's API usage.

Example Conversations

Finding Creators

User: Find me fitness micro-influencers in the US with at least 3% engagement rate and rates under $500 per post

Claude: I'll search for fitness creators matching your criteria.

[Uses search_creators tool]

I found 15 creators matching your criteria. Here are the top 5:

1. @FitWithSarah - 45K followers, 4.2% engagement, $350/post
2. @GymLifeMike - 62K followers, 3.8% engagement, $400/post
...

Running a Campaign

User: Create a summer fitness campaign with a $10,000 budget running from June to August

Claude: I'll create that campaign for you.

[Uses create_campaign tool]

Done! I've created "Summer Fitness Campaign" (cmp_abc123):
- Budget: $10,000
- Timeline: June 1 - August 31
- Status: Draft

Would you like me to activate it or search for creators to invite?

Discovering Creators

User: Find creators for my new skincare campaign

Claude: I'll trigger AI-powered discovery for your campaign.

[Uses trigger_discovery tool]
[Uses get_discovery_status tool]

Discovery found 32 potential creators. Here are the top matches:
...

Would you like me to bulk invite the top 10?

Managing Content Reviews

User: What content is pending review?

Claude: Let me check your pending reviews.

[Uses list_content tool with status="pending_review"]

You have 3 pieces of content awaiting review:

1. @FitWithSarah - Summer Protein Reel (submitted 2 hours ago)
2. @GymLifeMike - Workout Tips Post (submitted yesterday)
3. @YogaWithJen - Morning Routine Story (submitted 3 days ago)

Would you like me to show you any of these submissions?

Development

Running Locally

# Clone the repository
git clone https://github.com/srgirsky/InfluenceFlow.git
cd InfluenceFlow/influenceflow-mcp

# Install dependencies
pip install -e .

# Run the server
export INFLUENCEFLOW_API_KEY="your_key"
python -m influenceflow_mcp.server

Running Tests

pip install -e ".[dev]"
pytest

Environment Variables

License

MIT License - see LICENSE file for details.

Support

• Documentation: https://docs.influenceflow.co/mcp
• Issues: https://github.com/srgirsky/InfluenceFlow/issues
• Email: support@influenceflow.co