flowise-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FLOWISE_API_KEY | Yes | API key for authentication | |
| FLOWISE_TIMEOUT | No | Request timeout in seconds (default: 60) | 60 |
| FLOWISE_BASE_URL | Yes | Your Flowise instance URL |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tasks | {
"list": {},
"cancel": {},
"requests": {
"tools": {
"call": {}
},
"prompts": {
"get": {}
},
"resources": {
"read": {}
}
}
} |
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| flowise_list_flowsA | List all chatflows and agentflows in the Flowise instance. This tool retrieves a summary of all available flows, including their deployment status, visibility, and categories. Args: params: Input parameters containing optional flow_type filter and response_format. Returns: A formatted list of all flows with their basic information. Examples: - List all flows: Use with no parameters - List only agentflows: Use flow_type='AGENTFLOW' - Get JSON output: Use response_format='json' |
| flowise_get_flowA | Get detailed information about a specific chatflow or agentflow. This tool retrieves the full configuration of a flow, including all nodes, edges, and settings. Args: params: Input containing flow_id and response_format. Returns: Detailed flow information including configuration and nodes. Examples: - Get flow details: Use with the flow ID from flowise_list_flows |
| flowise_predictA | Send a message to a chatflow or agentflow and get a response. This is the primary tool for interacting with Flowise flows. It sends a question/message to the specified flow and returns the AI response. Args: params: Input containing flow_id, question, and optional session_id, streaming preference, and override_config. Returns: The response from the Flowise flow. Examples: - Simple question: Use flow_id and question - With session: Add session_id to maintain conversation context - Override settings: Use override_config to adjust temperature, etc. |
| flowise_analyze_flowA | Analyze a chatflow or agentflow and provide improvement suggestions. This tool examines the flow configuration and provides actionable recommendations for enhancing the flow's capabilities, performance, and best practices compliance. IMPORTANT: This is the primary tool for answering questions like "How can I improve this agentflow to do X?" or "What can I add to make my chatflow better at Y?" Args: params: Input containing flow_id, optional improvement_goal, and response_format. Returns: A detailed analysis with: - Current flow structure overview - Identified issues or gaps - Prioritized improvement suggestions - Best practices recommendations - Specific nodes to add or configure Examples: - General analysis: Use with just the flow_id - Targeted improvements: Add improvement_goal like "improve accuracy" - Speed optimization: Use improvement_goal="faster responses" - Add capabilities: Use improvement_goal="handle customer support queries" |
| flowise_create_flowA | Create a new chatflow or agentflow in Flowise. This tool creates a new flow with the specified configuration. The flow_data should be a valid JSON string containing the nodes and edges configuration. Args: params: Input containing name, flow_data (JSON), flow_type, is_public, and category. Returns: The created flow's details including its new ID. Examples: - Create a simple chatflow with a name and empty flow_data: '{}' - Create an agentflow: Set flow_type='AGENTFLOW' |
| flowise_update_flowA | Update an existing chatflow or agentflow. This tool updates the specified flow with new configuration. Only provided fields will be updated. Args: params: Input containing flow_id and optional fields to update. Returns: Confirmation of the update with the flow's details. Examples: - Rename a flow: Use flow_id and name - Update configuration: Use flow_id and flow_data - Make public: Use flow_id and is_public=True |
| flowise_delete_flowA | Delete a chatflow or agentflow from Flowise. WARNING: This action is irreversible. The flow and its configuration will be permanently deleted. Args: params: Input containing the flow_id to delete. Returns: Confirmation of deletion. |
| flowise_get_chat_historyA | Retrieve chat message history for a specific flow. This tool gets the conversation history from a chatflow or agentflow, useful for reviewing past interactions or debugging. Args: params: Input containing flow_id, optional session_id, limit, and response_format. Returns: List of chat messages with timestamps and content. |
| flowise_list_variablesA | List all global variables configured in Flowise. Variables can be used across flows for storing API keys, URLs, or other configuration values. Args: params: Input containing response_format. Returns: List of configured variables. |
| flowise_list_toolsA | List all tools available in Flowise. This retrieves the list of registered tools that can be used in agentflows and chatflows. Args: params: Input containing response_format. Returns: List of available tools with their descriptions. |
| flowise_pingA | Check if the Flowise server is reachable and responding. Use this tool to verify connectivity before making other requests. Returns: Server status message. |
| flowise_list_assistantsB | List all assistants configured in Flowise. Assistants are pre-configured AI agents that can be used for specific tasks. Args: params: Input containing response_format. Returns: List of configured assistants. |
| flowise_get_assistantB | Get detailed information about a specific assistant. Args: params: Input containing assistant_id and response_format. Returns: Detailed assistant information including configuration. |
| flowise_delete_chat_historyA | Delete chat message history for a specific flow. This removes conversation history from the database. Use with caution. Args: params: Input containing flow_id, optional session_id and chat_id. Returns: Confirmation of deletion. |
| flowise_list_document_storesA | List all document stores configured in Flowise. Document stores contain indexed documents for RAG (Retrieval-Augmented Generation). Args: params: Input containing response_format. Returns: List of document stores with their details. |
| flowise_get_document_storeB | Get detailed information about a specific document store. Args: params: Input containing store_id and response_format. Returns: Detailed document store information. |
| flowise_upsert_vectorB | Insert or update vectors in a chatflow's vector store. This triggers the flow's document processing pipeline to update the vector store. Args: params: Input containing flow_id, optional override_config and stop_node_id. Returns: Summary of the upsert operation results. |
| flowise_query_vector_storeA | Execute a retrieval query on a document store's vector store. This searches for relevant documents based on the query. Args: params: Input containing store_id and query. Returns: Retrieved documents with relevance information. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| analyze_agentflow | Prompt template for analyzing an agentflow and suggesting improvements. |
| improve_chatbot | Prompt template for improving a chatbot based on a specific issue. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| list_all_flows | Resource to get all flows as a JSON list. |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/JuliDir/flowise-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server