@kenkaiiii/flowforge-mcp
v1.0.14
Published
FlowForge MCP Server - Connect LLMs to FlowForge workflow automation platform
Maintainers
Readme
FlowForge MCP Server
Professional MCP server that connects LLMs to FlowForge workflow automation platform.
Installation
npm install -g flowforgeSetup
- Set your FlowForge API key:
export FLOWFORGE_API_KEY="ff_sk_your_api_key_here"
export FLOWFORGE_API_URL="http://localhost:8080" # Optional, defaults to localhost- Run the MCP server:
flowforge-mcpMCP Tools
Node Discovery
flowforge_discover_nodes- Get available node typesflowforge_get_node_schema- Get node configuration requirementsflowforge_get_node_examples- Get real-world config examples
Workflow Operations
flowforge_validate_workflow- Validate workflow before creationflowforge_test_workflow- Dry-run workflow executionflowforge_create_workflow- Create workflow from DSLflowforge_list_workflows- List existing workflowsflowforge_manage_workflow- Update/delete/test workflows
Example Usage
Create E-commerce Workflow
LLM: Use flowforge_create_workflow with:
dsl: "webhook -> ai_agent -> slack"
configs: {
"webhook": {"endpoint": "/orders"},
"ai_agent": {
"model": {"provider": "openai", "model_name": "gpt-4"},
"prompt": {"system_prompt": "Sales analyst", "user_prompt": "Analyze: {{webhook.data}}"}
},
"slack": {"channel": "#sales", "message": "Order: {{ai_agent.output}}"}
}Validate Before Creating
LLM: Use flowforge_validate_workflow to check configurationTest Workflow
LLM: Use flowforge_test_workflow with sample dataConfiguration
Required environment variables:
FLOWFORGE_API_KEY- Your FlowForge API keyFLOWFORGE_API_URL- FlowForge server URL (optional)
License
MIT
