{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github_mcp-leonardsellem-n8n-mcp-server","slug":"mcp-leonardsellem-n8n-mcp-server","name":"n8n-mcp-server","type":"mcp","url":"https://github.com/leonardsellem/n8n-mcp-server","page_url":"https://unfragile.ai/mcp-leonardsellem-n8n-mcp-server","categories":["mcp-servers","automation"],"tags":[],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_0","uri":"capability://tool.use.integration.mcp.native.workflow.crud.operations.with.structured.tool.definitions","name":"mcp-native workflow crud operations with structured tool definitions","description":"Exposes n8n workflow lifecycle management (create, read, update, delete) through the Model Context Protocol's tool system, using JSON schema-based tool definitions that allow AI assistants to invoke workflow operations with type-safe parameters. Each operation maps directly to n8n REST API endpoints (POST /workflows, GET /workflows/{id}, etc.) with automatic parameter validation and error handling at the MCP layer.","intents":["I want my AI assistant to programmatically create new n8n workflows based on user requirements","I need to let Claude or another LLM list and inspect existing workflows without direct API access","I want to enable workflow updates through natural language commands from an AI agent"],"best_for":["AI agent developers building autonomous workflow management systems","Teams integrating n8n with Claude, ChatGPT, or other MCP-compatible LLMs","Non-technical users who want AI-mediated workflow creation without API knowledge"],"limitations":["No built-in workflow validation before submission — relies on n8n API to reject invalid configurations","Workflow complexity is limited by n8n's API payload size constraints (typically 1-10MB per workflow definition)","No transactional guarantees — partial failures in multi-step workflow creation are not rolled back"],"requires":["Node.js 18 or later","n8n instance with API access enabled","Valid N8N_API_KEY environment variable","N8N_API_URL pointing to n8n API endpoint (e.g., http://localhost:5678/api/v1)"],"input_types":["JSON workflow definition objects","Workflow ID strings","Workflow name and description strings"],"output_types":["Workflow metadata objects (id, name, active status, createdAt)","Full workflow definitions with node configurations","Success/error status messages"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_1","uri":"capability://tool.use.integration.dual.mode.workflow.execution.via.api.and.webhook.triggers","name":"dual-mode workflow execution via api and webhook triggers","description":"Provides two distinct execution pathways for n8n workflows: direct API execution (execution_run tool) that triggers workflows synchronously through the n8n REST API, and webhook execution (run_webhook tool) that invokes workflows via HTTP webhook endpoints with optional basic authentication. The server abstracts both mechanisms through a unified tool interface, allowing AI assistants to choose execution mode based on workflow requirements (synchronous vs. asynchronous, authenticated vs. public).","intents":["I want to trigger a workflow synchronously and wait for results in my AI agent's decision loop","I need to invoke a workflow asynchronously via webhook without blocking the AI assistant","I want to execute workflows that require authentication through a secure webhook endpoint"],"best_for":["Developers building real-time AI agents that depend on workflow output","Teams with long-running workflows that should execute asynchronously","Organizations requiring webhook-based integrations with basic auth security"],"limitations":["API execution is synchronous only — no built-in timeout handling for long-running workflows (may block MCP server)","Webhook execution provides no built-in response polling — caller must implement their own status checking","Basic auth for webhooks is transmitted in HTTP headers — requires HTTPS in production to prevent credential exposure","No support for webhook signature verification — relies on basic auth alone for security"],"requires":["N8N_API_URL and N8N_API_KEY for API execution mode","N8N_WEBHOOK_USERNAME and N8N_WEBHOOK_PASSWORD for webhook execution mode (optional if webhooks are public)","Webhook endpoint URL accessible from the MCP server's network"],"input_types":["Workflow ID (string)","Execution parameters (JSON object)","Webhook URL (string)","Basic auth credentials (username/password)"],"output_types":["Execution ID (string)","Execution status (running, success, error)","Workflow output data (JSON)","HTTP response from webhook (text or JSON)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_10","uri":"capability://memory.knowledge.workflow.retrieval.with.full.configuration.inspection","name":"workflow retrieval with full configuration inspection","description":"Provides a tool to fetch complete workflow definitions (workflow_get) by workflow ID, returning the full configuration including all nodes, connections, credentials, and metadata. This allows AI assistants to inspect existing workflows, understand their structure, and use that information for modification or cloning. The tool returns the exact workflow definition that would be used for updates or exports.","intents":["I want to examine a workflow's structure before modifying it","I need to understand what nodes and connections exist in a workflow","I want my AI agent to clone or adapt an existing workflow"],"best_for":["AI agents that need to understand workflow structure before modification","Developers building workflow analysis or migration tools","Teams that want to version control or audit workflow configurations"],"limitations":["Returns complete workflow definition — may be very large for complex workflows, exceeding context window limits","No filtering or partial retrieval — must fetch entire workflow even if only specific nodes are needed","Credentials are included in the response but may be redacted by n8n API — cannot inspect actual credential values"],"requires":["N8N_API_KEY with read permissions","Workflow ID (string)","N8N_API_URL"],"input_types":["Workflow ID (string)"],"output_types":["Complete workflow definition object (JSON) with nodes, connections, metadata"],"categories":["memory-knowledge","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_11","uri":"capability://search.retrieval.workflow.listing.with.metadata.summary","name":"workflow listing with metadata summary","description":"Provides a tool to list all workflows in the n8n instance (workflow_list) with summary metadata including workflow ID, name, active status, creation date, and last update time. This allows AI assistants to discover available workflows, understand the workflow inventory, and select specific workflows for further operations. The list is returned as an array of workflow summary objects.","intents":["I want to see all workflows available in my n8n instance","I need to find a specific workflow by name or ID","I want my AI agent to understand the workflow inventory before making decisions"],"best_for":["Workflow discovery and inventory management","AI agents that need to understand available workflows before taking action","Teams with many workflows that need programmatic listing"],"limitations":["No filtering or search — returns all workflows, which may be large for instances with hundreds of workflows","No pagination — large workflow lists may exceed context window limits","Summary metadata only — does not include node details or execution history"],"requires":["N8N_API_KEY with read permissions","N8N_API_URL"],"input_types":[],"output_types":["Array of workflow summary objects (id, name, active, createdAt, updatedAt)"],"categories":["search-retrieval","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_2","uri":"capability://tool.use.integration.execution.monitoring.and.lifecycle.control.with.status.polling","name":"execution monitoring and lifecycle control with status polling","description":"Provides tools to query execution status (execution_get, execution_list), stop running executions (execution_stop), and retrieve execution statistics through the Resources System. The implementation polls the n8n API for execution state, allowing AI assistants to monitor workflow progress, detect failures, and make decisions based on execution outcomes without requiring webhooks or event subscriptions.","intents":["I want to check if a workflow execution has completed and retrieve its output","I need to list all executions for a workflow to find a specific run","I want to stop a long-running workflow execution that's taking too long"],"best_for":["AI agents that need to wait for workflow completion before proceeding","Monitoring systems that poll execution status periodically","Developers building interactive workflows where users can cancel long-running tasks"],"limitations":["Polling-based approach introduces latency — status updates are not real-time, typically 1-5 second delays","No built-in exponential backoff — rapid polling can overload n8n API if not rate-limited by caller","Execution history is limited by n8n's retention policy — old executions may be purged from the API","No streaming support — large execution outputs must be fetched as complete JSON objects"],"requires":["N8N_API_KEY with read permissions for executions","N8N_API_URL pointing to n8n instance","Workflow ID for listing executions"],"input_types":["Execution ID (string)","Workflow ID (string)","Query filters (optional: status, limit, offset)"],"output_types":["Execution object with status, startTime, endTime, output data","Array of execution objects","Execution statistics (total runs, success rate, average duration)","Success/error status for stop operations"],"categories":["tool-use-integration","automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_3","uri":"capability://memory.knowledge.static.and.dynamic.resource.exposure.via.mcp.resource.protocol","name":"static and dynamic resource exposure via mcp resource protocol","description":"Exposes n8n data as MCP resources (n8n://workflows/list, n8n://workflow/{id}, n8n://execution-stats, etc.), allowing AI assistants to retrieve structured information about workflows and executions as readable resources rather than tool outputs. Static resources (workflow list, health status) are fetched on-demand, while dynamic resources support parameterized queries (e.g., n8n://workflow/123 returns details for workflow 123). This enables AI assistants to reference n8n data in their context window without explicit tool invocations.","intents":["I want my AI assistant to have access to the current list of workflows without calling a tool","I need the AI to understand workflow structure by reading it as a resource","I want execution statistics available in the assistant's context for decision-making"],"best_for":["AI assistants that benefit from having n8n data in their system prompt or context","Developers building multi-turn conversations where workflow state should be continuously available","Teams using Claude with MCP where resources are preferred over tool calls for data retrieval"],"limitations":["Resources are read-only — no mutations through the resource protocol (must use tools)","Dynamic resources require exact IDs — no fuzzy matching or search within resource paths","Resource content is not paginated — large workflow lists may exceed context window limits","No caching — each resource read triggers an API call to n8n, potentially causing rate limiting"],"requires":["N8N_API_KEY with read permissions","N8N_API_URL pointing to n8n instance","MCP client that supports resource protocol (Claude, etc.)"],"input_types":["Resource URI (string, e.g., n8n://workflow/123)","Query parameters (optional, embedded in URI)"],"output_types":["JSON-formatted workflow metadata","Execution statistics objects","Health status information","Workflow definition objects"],"categories":["memory-knowledge","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_4","uri":"capability://automation.workflow.environment.driven.configuration.with.multi.instance.support","name":"environment-driven configuration with multi-instance support","description":"Manages n8n connection configuration through environment variables (N8N_API_URL, N8N_API_KEY, N8N_WEBHOOK_USERNAME, N8N_WEBHOOK_PASSWORD), allowing the MCP server to connect to different n8n instances by changing environment variables. The configuration is loaded at server startup and used to initialize API clients, supporting both local and remote n8n instances with optional webhook authentication. This enables deployment flexibility without code changes.","intents":["I want to deploy the MCP server against different n8n instances (dev, staging, prod) without rebuilding","I need to secure API credentials using environment variables rather than hardcoding them","I want to enable webhook authentication for my n8n instance"],"best_for":["DevOps teams deploying MCP servers across multiple environments","Organizations with strict credential management policies","Teams running both local and cloud-hosted n8n instances"],"limitations":["Configuration is static at server startup — changing environment variables requires server restart","No built-in credential rotation — API keys must be manually updated in environment","Webhook credentials are stored in plaintext in environment — requires secure environment variable management","No support for multiple n8n instances in a single server instance — one server = one n8n connection"],"requires":["Environment variables: N8N_API_URL, N8N_API_KEY","Optional: N8N_WEBHOOK_USERNAME, N8N_WEBHOOK_PASSWORD","Optional: DEBUG flag for verbose logging"],"input_types":["Environment variable strings"],"output_types":["Initialized API client configuration","Connection validation status"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_5","uri":"capability://automation.workflow.layered.error.handling.and.debug.logging.with.context.preservation","name":"layered error handling and debug logging with context preservation","description":"Implements error handling at multiple layers (MCP protocol layer, n8n API layer, transport layer) with optional debug logging controlled by the DEBUG environment variable. Errors from n8n API calls are caught, transformed into MCP-compatible error responses, and logged with context (request parameters, API response status). This allows AI assistants to understand why operations failed and enables developers to diagnose issues through server logs.","intents":["I want my AI assistant to understand why a workflow creation failed and suggest fixes","I need to debug why the MCP server isn't connecting to my n8n instance","I want detailed logs of all API calls for auditing and troubleshooting"],"best_for":["Developers debugging MCP server integration issues","Teams requiring audit trails of all workflow operations","Organizations with strict logging and compliance requirements"],"limitations":["Debug logging is global — cannot selectively enable logging for specific operations","Error messages from n8n API are passed through without sanitization — may expose internal details","No structured logging format — logs are plain text, not JSON, making parsing difficult","Logs are written to stdout/stderr — requires external log aggregation for persistent storage"],"requires":["DEBUG environment variable set to 'true' for verbose logging","Access to server stdout/stderr for viewing logs"],"input_types":["DEBUG flag (boolean)"],"output_types":["Error messages with context","Debug log entries","Stack traces (when DEBUG is enabled)"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_6","uri":"capability://automation.workflow.workflow.activation.and.deactivation.state.management","name":"workflow activation and deactivation state management","description":"Provides tools to activate (workflow_activate) and deactivate (workflow_deactivate) n8n workflows, controlling whether workflows respond to triggers. These operations toggle the workflow's active status through the n8n API, allowing AI assistants to enable/disable workflows without modifying their configuration. The state change is immediate and persisted in n8n's database.","intents":["I want to temporarily disable a workflow without deleting it","I need to activate a workflow that was previously disabled","I want my AI agent to manage workflow availability based on business rules"],"best_for":["Workflow operators who need to pause workflows without losing configuration","AI agents managing workflow availability based on time or conditions","Teams with workflows that should be toggled on/off based on external events"],"limitations":["No graceful shutdown — deactivating a workflow stops it immediately, potentially interrupting running executions","No state validation — cannot check if a workflow is already active before activating it","No bulk operations — must activate/deactivate workflows one at a time"],"requires":["N8N_API_KEY with write permissions","Workflow ID (string)","N8N_API_URL"],"input_types":["Workflow ID (string)"],"output_types":["Updated workflow object with active status","Success/error status"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_7","uri":"capability://automation.workflow.workflow.deletion.with.permanent.removal","name":"workflow deletion with permanent removal","description":"Provides a tool to permanently delete workflows (workflow_delete) from the n8n instance through the REST API. This operation removes the workflow definition and all associated metadata, making it irreversible. The tool requires explicit workflow ID specification to prevent accidental deletions.","intents":["I want to remove a workflow that's no longer needed","I need my AI agent to clean up test workflows after validation","I want to delete workflows that failed configuration"],"best_for":["Workflow cleanup and maintenance operations","AI agents that create temporary workflows and need to remove them","Teams with automated workflow lifecycle management"],"limitations":["Deletion is permanent and irreversible — no undo or recovery mechanism","No confirmation prompt — deletion happens immediately upon tool invocation","No cascade deletion — dependent workflows or integrations are not automatically cleaned up","Execution history for deleted workflows may be retained in n8n, creating orphaned records"],"requires":["N8N_API_KEY with delete permissions","Workflow ID (string)","N8N_API_URL"],"input_types":["Workflow ID (string)"],"output_types":["Success/error status","Confirmation message"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_8","uri":"capability://automation.workflow.workflow.update.with.configuration.modification","name":"workflow update with configuration modification","description":"Provides a tool to update existing workflow definitions (workflow_update) by submitting modified workflow configuration objects to the n8n API. The tool accepts a workflow ID and a partial or complete workflow definition, merging changes with the existing configuration. This allows AI assistants to modify workflow nodes, connections, and settings without recreating the entire workflow.","intents":["I want to modify a workflow's node configuration based on new requirements","I need to update workflow settings like name, description, or tags","I want my AI agent to adjust workflow logic in response to user feedback"],"best_for":["Developers building AI-driven workflow modification systems","Teams with workflows that need dynamic reconfiguration","AI agents that learn and adapt workflows based on execution results"],"limitations":["No schema validation before submission — invalid workflow definitions are rejected by n8n API","No diff or preview — changes are applied immediately without showing what changed","No rollback mechanism — failed updates may leave workflows in inconsistent state","Complex workflow modifications (node rewiring, credential updates) require deep knowledge of n8n's workflow schema"],"requires":["N8N_API_KEY with write permissions","Workflow ID (string)","Updated workflow definition (JSON object)","N8N_API_URL"],"input_types":["Workflow ID (string)","Workflow definition object (JSON)"],"output_types":["Updated workflow object with new configuration","Success/error status"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github_mcp-leonardsellem-n8n-mcp-server__cap_9","uri":"capability://automation.workflow.workflow.creation.with.node.based.configuration","name":"workflow creation with node-based configuration","description":"Provides a tool to create new workflows (workflow_create) by accepting a workflow definition object containing nodes, connections, and metadata. The tool submits the definition to the n8n API, which validates the structure and creates the workflow in the database. This allows AI assistants to generate complete workflows from scratch based on user requirements, including node configuration, data flow, and trigger setup.","intents":["I want my AI assistant to generate a workflow that integrates multiple APIs based on user description","I need to create workflows programmatically without manual UI interaction","I want AI agents to build workflows that transform data through a series of nodes"],"best_for":["AI agents that generate workflows from natural language descriptions","Low-code platforms that use n8n as a backend workflow engine","Teams automating workflow creation for repetitive integration patterns"],"limitations":["Requires deep knowledge of n8n's workflow schema — node types, properties, connection format","No template system — each workflow must be defined from scratch or use external templates","No validation before submission — invalid schemas are rejected by n8n API with generic error messages","Complex workflows with many nodes may exceed payload size limits or timeout during creation"],"requires":["N8N_API_KEY with write permissions","Valid workflow definition object conforming to n8n schema","N8N_API_URL"],"input_types":["Workflow definition object (JSON) containing nodes, connections, metadata"],"output_types":["Created workflow object with ID, name, and metadata","Success/error status with validation errors if applicable"],"categories":["automation-workflow","tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":41,"verified":false,"data_access_risk":"high","permissions":["Node.js 18 or later","n8n instance with API access enabled","Valid N8N_API_KEY environment variable","N8N_API_URL pointing to n8n API endpoint (e.g., http://localhost:5678/api/v1)","N8N_API_URL and N8N_API_KEY for API execution mode","N8N_WEBHOOK_USERNAME and N8N_WEBHOOK_PASSWORD for webhook execution mode (optional if webhooks are public)","Webhook endpoint URL accessible from the MCP server's network","N8N_API_KEY with read permissions","Workflow ID (string)","N8N_API_URL"],"failure_modes":["No built-in workflow validation before submission — relies on n8n API to reject invalid configurations","Workflow complexity is limited by n8n's API payload size constraints (typically 1-10MB per workflow definition)","No transactional guarantees — partial failures in multi-step workflow creation are not rolled back","API execution is synchronous only — no built-in timeout handling for long-running workflows (may block MCP server)","Webhook execution provides no built-in response polling — caller must implement their own status checking","Basic auth for webhooks is transmitted in HTTP headers — requires HTTPS in production to prevent credential exposure","No support for webhook signature verification — relies on basic auth alone for security","Returns complete workflow definition — may be very large for complex workflows, exceeding context window limits","No filtering or partial retrieval — must fetch entire workflow even if only specific nodes are needed","Credentials are included in the response but may be redacted by n8n API — cannot inspect actual credential values","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.4913092797761881,"quality":0.34,"ecosystem":0.49999999999999994,"match_graph":0.25,"freshness":0.6,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.15,"match_graph":0.23,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:22.065Z","last_scraped_at":"2026-05-03T14:23:44.761Z","last_commit":"2025-07-09T21:05:08Z"},"community":{"stars":1610,"forks":289,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=mcp-leonardsellem-n8n-mcp-server","compare_url":"https://unfragile.ai/compare?artifact=mcp-leonardsellem-n8n-mcp-server"}},"signature":"qnYA7XApaQFPLKpTMgBAHtpWNA7M/RQ3inz5S2gTvf3Ro10JbpHi4KOvVYbl1larlOzS8rz6S/8XQnO/tak2Cg==","signedAt":"2026-06-22T02:22:18.796Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/mcp-leonardsellem-n8n-mcp-server","artifact":"https://unfragile.ai/mcp-leonardsellem-n8n-mcp-server","verify":"https://unfragile.ai/api/v1/verify?slug=mcp-leonardsellem-n8n-mcp-server","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}