{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"smithery_86755crj-n8n","slug":"86755crj-n8n","name":"n8n","type":"mcp","url":"https://smithery.ai/servers/86755crj/n8n","page_url":"https://unfragile.ai/86755crj-n8n","categories":["mcp-servers","automation"],"tags":["mcp","model-context-protocol","smithery:86755crj/n8n"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"smithery_86755crj-n8n__cap_0","uri":"capability://tool.use.integration.workflow.orchestration.via.mcp.protocol","name":"workflow-orchestration-via-mcp-protocol","description":"Exposes n8n's workflow execution engine through the Model Context Protocol, allowing LLM agents to trigger, monitor, and manage automation workflows as remote procedure calls. Uses MCP's standardized request/response format to abstract n8n's REST API, enabling stateless workflow invocation with parameter binding and execution state tracking without direct HTTP knowledge.","intents":["I want my AI agent to trigger complex multi-step automations based on LLM decisions","I need to invoke n8n workflows from Claude or other MCP-compatible LLMs without writing custom API clients","I want to chain LLM reasoning with deterministic workflow execution in a single agent loop"],"best_for":["AI agent developers building agentic workflows that need deterministic automation","teams integrating LLMs with existing n8n automation infrastructure","builders prototyping AI-driven business process automation"],"limitations":["MCP protocol adds ~100-300ms latency per workflow invocation due to serialization overhead","no built-in workflow versioning or rollback through MCP — must manage via n8n directly","workflow execution context (variables, secrets) must be pre-configured in n8n; cannot be dynamically injected via MCP","no real-time streaming of workflow execution logs — only final execution status available"],"requires":["n8n instance running (self-hosted or cloud) with API access enabled","MCP-compatible LLM client (Claude, or custom MCP client implementation)","n8n API credentials (API key or OAuth token)","network connectivity between MCP client and n8n instance"],"input_types":["workflow ID or name (string)","workflow input parameters (JSON object)","execution options (async/sync flag, timeout)"],"output_types":["execution ID (string)","execution status (running/success/error)","workflow output data (JSON object)","execution metadata (duration, timestamp)"],"categories":["tool-use-integration","workflow-automation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_1","uri":"capability://tool.use.integration.workflow.discovery.and.introspection","name":"workflow-discovery-and-introspection","description":"Queries n8n's workflow registry to enumerate available workflows, retrieve workflow definitions, and expose workflow input/output schemas through MCP's resource listing and metadata endpoints. Implements schema extraction from n8n's internal workflow graph representation, allowing LLM agents to discover executable workflows and understand their expected parameters without manual documentation.","intents":["I want my LLM agent to discover what workflows are available in our n8n instance","I need to expose workflow input schemas so the agent knows what parameters to pass","I want to dynamically list workflows by tag or category for conditional agent routing"],"best_for":["multi-workflow environments where agents need dynamic workflow selection","teams with large n8n workflow libraries requiring programmatic discovery","builders implementing agent-driven workflow routing based on user intent"],"limitations":["schema extraction only covers explicitly defined workflow inputs — computed or dynamic fields may not be discoverable","no filtering by workflow status (active/inactive) — all workflows returned regardless of deployment state","workflow descriptions must be manually maintained in n8n; no auto-generation from workflow logic","discovery is read-only — cannot create or modify workflows through MCP"],"requires":["n8n instance with API access and authentication","sufficient API permissions to list workflows and read workflow definitions","MCP client with resource listing capability"],"input_types":["optional filter parameters (workflow name, tag, status)"],"output_types":["workflow list (array of workflow metadata)","workflow definition (JSON schema with inputs/outputs)","workflow tags and categories (string array)"],"categories":["tool-use-integration","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_2","uri":"capability://tool.use.integration.execution.status.polling.and.monitoring","name":"execution-status-polling-and-monitoring","description":"Implements polling-based execution monitoring through MCP, allowing agents to query workflow execution status, retrieve execution results, and track long-running workflows without blocking. Uses n8n's execution history API to fetch status updates, with configurable polling intervals and timeout handling to prevent agent deadlock on slow workflows.","intents":["I want my agent to wait for a workflow to complete and retrieve its results","I need to check if a long-running workflow is still executing or has failed","I want to implement retry logic if a workflow execution fails"],"best_for":["agents orchestrating multi-step workflows with dependencies","systems requiring execution result retrieval for downstream processing","builders implementing fault-tolerant automation chains"],"limitations":["polling-based approach adds latency — no real-time execution webhooks through MCP","no built-in exponential backoff — polling interval must be manually configured","execution history retention depends on n8n configuration — old executions may be pruned","no streaming of execution logs — only final status and output available"],"requires":["n8n instance with execution history API enabled","execution ID from a prior workflow invocation","polling timeout configuration (recommended 30-300 seconds)"],"input_types":["execution ID (string)","polling interval (milliseconds, optional)","timeout duration (seconds, optional)"],"output_types":["execution status (running/success/error/timeout)","execution result data (JSON object)","error details (string, if failed)","execution duration (milliseconds)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_3","uri":"capability://tool.use.integration.parameter.binding.and.context.injection","name":"parameter-binding-and-context-injection","description":"Enables dynamic parameter passing to workflows through MCP tool invocation, mapping LLM-generated parameters to n8n workflow input variables. Implements type coercion and validation against workflow input schemas, allowing agents to pass structured data (JSON objects, arrays) and primitive types directly into workflow execution contexts without manual serialization.","intents":["I want to pass data extracted by the LLM (like user input or API responses) into n8n workflows","I need to dynamically set workflow variables based on agent reasoning","I want type-safe parameter passing with validation before workflow execution"],"best_for":["agents that need to pass LLM-generated data into deterministic workflows","multi-step automation chains where LLM output feeds into workflow input","builders implementing data transformation pipelines with LLM + n8n"],"limitations":["parameter validation only covers schema-defined inputs — custom validation logic in n8n is not enforced through MCP","no support for file uploads or binary data — only JSON-serializable types","secrets and sensitive data passed as parameters are logged in n8n execution history — no automatic masking","parameter size limited by MCP message size constraints (typically 100MB)"],"requires":["workflow input schema defined in n8n","parameter values matching expected types (string, number, boolean, object, array)","MCP client capable of passing structured JSON data"],"input_types":["workflow input parameters (JSON object)","primitive types (string, number, boolean)","nested objects and arrays"],"output_types":["parameter validation result (success/error)","execution ID with parameters bound"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_4","uri":"capability://tool.use.integration.error.handling.and.execution.failure.reporting","name":"error-handling-and-execution-failure-reporting","description":"Captures workflow execution errors and exposes them through MCP as structured error objects, including error type, message, and failure context (which node failed, error code). Allows agents to distinguish between transient failures (retry-able) and permanent errors (configuration issues) based on error classification from n8n's execution logs.","intents":["I want my agent to detect when a workflow fails and understand why","I need to implement conditional retry logic based on error type","I want to log workflow failures for debugging and monitoring"],"best_for":["agents implementing fault-tolerant automation with retry strategies","systems requiring detailed error reporting for observability","builders debugging complex multi-workflow orchestrations"],"limitations":["error classification depends on n8n node implementations — custom nodes may not provide structured error data","no stack traces or detailed debugging information — only high-level error messages","error context limited to final failure point — intermediate node failures not exposed","no automatic error recovery suggestions — agents must implement retry logic"],"requires":["n8n instance with error logging enabled","execution ID from a failed workflow","MCP client capable of parsing structured error responses"],"input_types":["execution ID (string)"],"output_types":["error type (string, e.g., 'network_error', 'validation_error', 'timeout')","error message (string)","failing node name (string)","error code (string or number)","error context (JSON object with additional details)"],"categories":["tool-use-integration","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_5","uri":"capability://search.retrieval.workflow.execution.history.retrieval","name":"workflow-execution-history-retrieval","description":"Queries n8n's execution history database through MCP to retrieve past workflow executions, including execution timestamps, status, duration, and output data. Implements pagination and filtering by date range or status, allowing agents to audit workflow behavior, retrieve historical results, or implement idempotency checks based on prior executions.","intents":["I want to check if a workflow has already been executed with the same parameters (idempotency)","I need to retrieve historical execution data for reporting or analysis","I want to audit which workflows have failed in the last 24 hours"],"best_for":["agents implementing idempotent workflow execution","systems requiring execution audit trails and compliance logging","builders analyzing workflow performance and failure patterns"],"limitations":["execution history retention depends on n8n configuration — old executions may be automatically purged","no full-text search of execution logs — filtering limited to metadata (status, date, workflow ID)","pagination may be slow for large execution histories (millions of records)","no real-time streaming — history queries return static snapshots"],"requires":["n8n instance with execution history API enabled","sufficient API permissions to read execution history","optional: date range or status filters"],"input_types":["workflow ID (string)","date range (start/end timestamps, optional)","status filter (running/success/error, optional)","pagination parameters (limit, offset)"],"output_types":["execution list (array of execution records)","execution metadata (timestamp, duration, status)","execution output data (JSON object)","pagination info (total count, has_more)"],"categories":["search-retrieval","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_6","uri":"capability://planning.reasoning.multi.workflow.orchestration.and.chaining","name":"multi-workflow-orchestration-and-chaining","description":"Enables sequential or conditional execution of multiple n8n workflows through MCP, allowing agents to chain workflow outputs as inputs to subsequent workflows. Implements execution dependency tracking and conditional branching based on prior workflow results, enabling complex multi-step automation scenarios without manual workflow composition in n8n.","intents":["I want to execute workflow A, then pass its output to workflow B","I need to conditionally execute workflow C only if workflow B succeeds","I want to orchestrate a multi-step process where each step depends on the previous result"],"best_for":["agents implementing complex multi-step business processes","systems requiring conditional workflow routing based on intermediate results","builders orchestrating data pipelines with multiple transformation stages"],"limitations":["no built-in transaction semantics — partial failures in multi-workflow chains require manual rollback logic","execution order must be explicitly managed by agent — no automatic dependency resolution","no distributed tracing across workflows — debugging multi-workflow chains requires manual log correlation","timeout handling is per-workflow, not per-chain — long chains may exceed agent timeout limits"],"requires":["multiple n8n workflows with compatible input/output schemas","MCP client capable of sequential tool invocation","agent logic to manage execution order and data flow"],"input_types":["workflow execution sequence (array of workflow IDs)","conditional logic (if-then rules based on prior results)","data transformation rules (mapping prior output to next input)"],"output_types":["final workflow output (JSON object)","execution chain metadata (all execution IDs, durations)","intermediate results (optional, for debugging)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"smithery_86755crj-n8n__cap_7","uri":"capability://safety.moderation.credential.and.secret.management.abstraction","name":"credential-and-secret-management-abstraction","description":"Abstracts n8n's credential storage system through MCP, allowing agents to reference pre-configured credentials (API keys, database passwords, OAuth tokens) by name without exposing secrets in agent prompts or logs. Uses n8n's credential encryption and access control to ensure secrets remain secure while enabling workflows to access external services.","intents":["I want workflows to use stored credentials without the agent knowing the actual secret values","I need to ensure API keys and passwords are never exposed in agent logs or prompts","I want to rotate credentials in n8n without updating agent code"],"best_for":["agents executing workflows that require external API access","systems with strict security requirements (no secrets in logs)","teams managing multiple credentials across workflows"],"limitations":["agents cannot create or rotate credentials through MCP — must be managed in n8n UI","no fine-grained credential access control — agents can reference any credential available to their n8n user","credential names must be known in advance — no discovery of available credentials through MCP","no audit logging of credential usage through MCP — only n8n's internal audit trail"],"requires":["credentials pre-configured in n8n instance","credential names known to agent or discoverable through workflow introspection","n8n user with permission to use referenced credentials"],"input_types":["credential name (string)"],"output_types":["credential reference (opaque token for workflow use)","credential type (string, e.g., 'api_key', 'oauth2', 'database')"],"categories":["safety-moderation","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":27,"verified":false,"data_access_risk":"high","permissions":["n8n instance running (self-hosted or cloud) with API access enabled","MCP-compatible LLM client (Claude, or custom MCP client implementation)","n8n API credentials (API key or OAuth token)","network connectivity between MCP client and n8n instance","n8n instance with API access and authentication","sufficient API permissions to list workflows and read workflow definitions","MCP client with resource listing capability","n8n instance with execution history API enabled","execution ID from a prior workflow invocation","polling timeout configuration (recommended 30-300 seconds)"],"failure_modes":["MCP protocol adds ~100-300ms latency per workflow invocation due to serialization overhead","no built-in workflow versioning or rollback through MCP — must manage via n8n directly","workflow execution context (variables, secrets) must be pre-configured in n8n; cannot be dynamically injected via MCP","no real-time streaming of workflow execution logs — only final execution status available","schema extraction only covers explicitly defined workflow inputs — computed or dynamic fields may not be discoverable","no filtering by workflow status (active/inactive) — all workflows returned regardless of deployment state","workflow descriptions must be manually maintained in n8n; no auto-generation from workflow logic","discovery is read-only — cannot create or modify workflows through MCP","polling-based approach adds latency — no real-time execution webhooks through MCP","no built-in exponential backoff — polling interval must be manually configured","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.26,"ecosystem":0.49000000000000005,"match_graph":0.25,"freshness":0.5,"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":"2026-05-24T12:16:25.061Z","last_scraped_at":"2026-05-03T15:19:41.319Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=86755crj-n8n","compare_url":"https://unfragile.ai/compare?artifact=86755crj-n8n"}},"signature":"g+jGFqc74sB/wMZznEC0yVM7AADkUvtO8Yk++D/zoqEZJcU3/haYF8m15yU115gy6iRbe6bygx3fQRzFk8L+AQ==","signedAt":"2026-06-21T09:21:07.498Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/86755crj-n8n","artifact":"https://unfragile.ai/86755crj-n8n","verify":"https://unfragile.ai/api/v1/verify?slug=86755crj-n8n","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"}}