{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"npm-n8n-mcp","slug":"n8n-mcp","name":"n8n-mcp","type":"mcp","url":"https://github.com/czlonkowski/n8n-mcp#readme","page_url":"https://unfragile.ai/n8n-mcp","categories":["mcp-servers","automation"],"tags":["n8n","mcp","model-context-protocol","ai","workflow","automation"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"npm-n8n-mcp__cap_0","uri":"capability://tool.use.integration.n8n.workflow.exposure.via.mcp.server.interface","name":"n8n workflow exposure via mcp server interface","description":"Exposes n8n workflow automation capabilities as MCP server resources, allowing Claude and other MCP clients to discover and invoke n8n workflows through a standardized protocol. Implements MCP server specification with resource listing endpoints that map n8n workflows to callable tools, enabling AI agents to treat n8n as a composable backend service without direct API knowledge.","intents":["Enable Claude to trigger n8n workflows as part of multi-step reasoning chains","Expose n8n automation capabilities to AI agents without building custom API wrappers","Allow LLM-driven agents to orchestrate complex n8n workflows as atomic operations"],"best_for":["AI agent developers building multi-tool orchestration systems","Teams integrating n8n automation with Claude or other MCP-compatible LLMs","Builders creating AI-augmented workflow platforms"],"limitations":["Requires n8n instance to be running and accessible from MCP server host","Workflow discovery is static at server startup — runtime workflow creation not reflected without restart","No built-in authentication caching — each MCP call must authenticate with n8n credentials"],"requires":["n8n instance (self-hosted or cloud) with API access enabled","MCP client implementation (Claude Desktop, custom MCP host, etc.)","Node.js 16+ runtime for MCP server process","n8n API credentials (API key or webhook authentication)"],"input_types":["MCP resource requests (JSON-RPC)","Workflow identifiers (n8n workflow IDs)","Tool invocation parameters (JSON)"],"output_types":["MCP resource listings (workflow metadata)","Workflow execution results (JSON)","Execution status and logs"],"categories":["tool-use-integration","workflow-orchestration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_1","uri":"capability://data.processing.analysis.dynamic.workflow.parameter.mapping.and.execution","name":"dynamic workflow parameter mapping and execution","description":"Translates MCP tool invocation parameters into n8n workflow input variables, executes the workflow with those parameters, and maps execution results back to MCP response format. Implements parameter schema inference from n8n workflow definitions to enable type-safe, context-aware parameter passing from AI agents to workflows without manual schema definition.","intents":["Pass Claude-generated parameters (e.g., email addresses, file paths) directly into n8n workflows","Execute parameterized n8n workflows with dynamic inputs from AI reasoning chains","Enable workflows to receive structured data from LLM outputs without manual formatting"],"best_for":["Developers building AI agents that need to pass computed values into n8n workflows","Teams automating multi-step processes where LLM reasoning feeds workflow inputs","Builders creating no-code automation platforms augmented with AI decision-making"],"limitations":["Parameter mapping assumes n8n workflow input variables are pre-defined — dynamic variable creation not supported","No type coercion — parameter types must match workflow expectations or execution fails","Large parameter payloads (>1MB) may exceed MCP message size limits","Workflow execution is synchronous — long-running workflows block MCP response until completion"],"requires":["n8n workflows with explicitly defined input variables","MCP client capable of passing structured JSON parameters","n8n API access with workflow execution permissions"],"input_types":["JSON parameters from MCP tool invocation","Workflow variable definitions (inferred from n8n)","Execution context metadata"],"output_types":["Workflow execution results (JSON)","Execution metadata (duration, status, error messages)","Structured output from n8n nodes"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_2","uri":"capability://safety.moderation.n8n.credential.management.and.secure.authentication","name":"n8n credential management and secure authentication","description":"Manages authentication between the MCP server and n8n instance, supporting multiple credential types (API keys, OAuth tokens, basic auth) with secure storage and injection into workflow execution contexts. Implements credential isolation so workflows can access n8n-stored credentials without exposing them to the MCP client, enabling secure multi-tenant workflow execution.","intents":["Authenticate MCP server to n8n without hardcoding credentials in configuration","Allow workflows to use n8n-stored credentials (database passwords, API keys) securely","Support multiple n8n instances or credential sets without credential duplication"],"best_for":["Teams running n8n in production with strict credential management policies","Multi-tenant AI agent platforms where workflows need isolated credential access","Builders integrating n8n with sensitive data sources (databases, payment APIs)"],"limitations":["Credential types supported are limited to n8n's native credential system — custom credential types not supported","No credential rotation automation — manual credential updates required in n8n","Credentials are transmitted over HTTP if n8n instance is not HTTPS-enabled","MCP server must have network access to n8n — no support for air-gapped deployments"],"requires":["n8n instance with credential management enabled","API key or OAuth token with credential access permissions","Secure credential storage (environment variables or secrets manager)"],"input_types":["Credential identifiers (n8n credential IDs)","Authentication tokens (API keys, OAuth tokens)","Credential type specifications"],"output_types":["Authentication headers for n8n API calls","Credential metadata (type, scope, expiration)","Execution context with injected credentials"],"categories":["safety-moderation","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_3","uri":"capability://memory.knowledge.workflow.discovery.and.metadata.introspection","name":"workflow discovery and metadata introspection","description":"Queries n8n API to enumerate available workflows, extract metadata (name, description, input/output schemas), and expose them as MCP resources with discoverable tool definitions. Implements caching of workflow metadata to reduce API calls while maintaining eventual consistency with n8n's workflow catalog.","intents":["Allow Claude to discover what n8n workflows are available without manual documentation","Enable AI agents to introspect workflow input/output schemas for parameter validation","Provide workflow descriptions to LLMs for intelligent workflow selection in multi-tool scenarios"],"best_for":["AI agent builders who want dynamic workflow discovery without hardcoding tool definitions","Teams with large n8n workflow catalogs that change frequently","Builders creating AI-augmented workflow platforms with self-documenting capabilities"],"limitations":["Metadata caching introduces staleness — newly created workflows not visible until cache refresh","Workflow descriptions must be manually maintained in n8n — no auto-generation from node structure","Complex workflows with conditional branches may have ambiguous input/output schemas","No support for workflow versioning — only latest version is exposed"],"requires":["n8n API access with workflow list and read permissions","Periodic cache refresh mechanism (polling or webhook)","Sufficient memory for metadata caching (typically <10MB for 1000+ workflows)"],"input_types":["n8n API responses (workflow list, workflow details)","Cache invalidation signals"],"output_types":["MCP resource listings (workflow metadata)","Tool definitions with input/output schemas","Workflow descriptions and documentation"],"categories":["memory-knowledge","search-retrieval"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_4","uri":"capability://automation.workflow.workflow.execution.result.streaming.and.status.tracking","name":"workflow execution result streaming and status tracking","description":"Monitors n8n workflow execution progress, streams intermediate results and logs back to the MCP client, and provides execution status updates (running, completed, failed) with error details. Implements polling or webhook-based status tracking to enable long-running workflow visibility without blocking MCP responses.","intents":["Track n8n workflow execution progress from Claude or other AI agents in real-time","Stream workflow logs and intermediate results for debugging and transparency","Enable AI agents to react to workflow failures with error context and retry logic"],"best_for":["Builders creating observable AI-augmented workflow platforms","Teams running long-running n8n workflows that need progress visibility","AI agent systems that need to handle workflow failures gracefully"],"limitations":["Polling-based status tracking adds latency (typically 1-5 second polling intervals)","Webhook-based tracking requires n8n to have outbound network access to MCP server","Large execution logs may exceed MCP message size limits — truncation required","No support for streaming partial results from individual workflow nodes"],"requires":["n8n API access with execution history and logs permissions","Either polling mechanism or webhook endpoint for status updates","Network connectivity between n8n and MCP server (for webhooks)"],"input_types":["Workflow execution IDs","Status polling requests","Webhook payloads from n8n"],"output_types":["Execution status updates (running, completed, failed)","Workflow logs and error messages","Intermediate execution results","Execution metadata (duration, node execution times)"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_5","uri":"capability://safety.moderation.error.handling.and.execution.failure.recovery","name":"error handling and execution failure recovery","description":"Captures n8n workflow execution errors, maps them to structured error responses, and provides retry logic with exponential backoff. Implements error classification (transient vs permanent) to enable intelligent retry strategies and error context propagation to MCP clients for AI-driven error handling.","intents":["Provide detailed error context to Claude for intelligent error recovery decisions","Automatically retry transient failures (network timeouts, rate limits) without AI intervention","Enable AI agents to implement custom error handling logic based on failure type"],"best_for":["Builders creating resilient AI agent systems that handle workflow failures gracefully","Teams running mission-critical n8n workflows that need automatic recovery","AI systems that need to learn from and adapt to workflow failures"],"limitations":["Retry logic is limited to MCP server layer — n8n's native error handling not integrated","Error classification heuristics may misclassify failures (e.g., permanent API errors as transient)","Exponential backoff may exceed workflow timeout constraints for long-running workflows","No support for custom error handlers or recovery workflows"],"requires":["n8n API access with execution error details","Configurable retry policy (max retries, backoff strategy)","Error classification rules or heuristics"],"input_types":["n8n execution error responses","Retry policy configuration","Error classification rules"],"output_types":["Structured error responses (error type, message, context)","Retry metadata (attempt count, next retry time)","Error classification (transient, permanent, unknown)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_6","uri":"capability://automation.workflow.multi.workflow.orchestration.and.chaining","name":"multi-workflow orchestration and chaining","description":"Enables sequential or conditional execution of multiple n8n workflows based on previous execution results, implementing workflow composition patterns (fan-out, fan-in, conditional branching) at the MCP layer. Allows AI agents to orchestrate complex multi-workflow processes by treating workflow chains as single MCP operations.","intents":["Chain multiple n8n workflows together based on AI reasoning (e.g., validate data, then process, then notify)","Implement conditional workflow execution based on intermediate results","Enable AI agents to compose complex automation sequences from simple n8n workflows"],"best_for":["Builders creating AI-driven process automation platforms","Teams with modular n8n workflows that need intelligent composition","AI agent systems that need to orchestrate multi-step workflows dynamically"],"limitations":["Workflow chaining is stateless — no persistent state between workflow executions","Conditional branching logic must be defined in MCP server code — not configurable via n8n UI","Complex orchestration patterns may exceed MCP message size limits","No support for parallel workflow execution — only sequential or conditional patterns"],"requires":["Multiple n8n workflows with compatible input/output schemas","Orchestration logic definition (sequential, conditional, fan-out patterns)","State management for intermediate results (in-memory or external store)"],"input_types":["Workflow chain definitions (sequence of workflow IDs)","Conditional logic rules","Initial input parameters"],"output_types":["Final workflow execution results","Execution history (all intermediate results)","Orchestration metadata (execution path, timing)"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_7","uri":"capability://tool.use.integration.mcp.client.compatibility.and.protocol.compliance","name":"mcp client compatibility and protocol compliance","description":"Implements the Model Context Protocol specification, enabling compatibility with any MCP-compliant client (Claude Desktop, custom MCP hosts, LLM frameworks). Handles MCP message serialization, resource discovery, tool invocation, and error responses according to the MCP standard.","intents":["Use n8n workflows with Claude Desktop without custom integration code","Integrate n8n with any MCP-compatible LLM framework or agent platform","Enable n8n workflows to be used as tools in multi-tool AI systems"],"best_for":["Developers building AI agents with MCP-compatible frameworks","Teams using Claude Desktop and wanting n8n integration","Builders creating multi-tool AI systems with standardized interfaces"],"limitations":["MCP protocol version compatibility — older clients may not support latest features","Message size limits (typically 1-10MB) constrain large workflow results","No support for streaming large results — must fit within single MCP message","Protocol overhead adds ~50-100ms latency per MCP call"],"requires":["MCP-compatible client (Claude Desktop, custom MCP host, etc.)","Node.js 16+ runtime for MCP server","Network connectivity between client and MCP server"],"input_types":["MCP protocol messages (JSON-RPC)","Resource requests","Tool invocation requests"],"output_types":["MCP protocol responses","Resource listings","Tool execution results"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_8","uri":"capability://safety.moderation.workflow.input.validation.and.schema.enforcement","name":"workflow input validation and schema enforcement","description":"Validates MCP tool invocation parameters against n8n workflow input schemas before execution, rejecting invalid inputs with detailed validation errors. Implements JSON schema validation with type coercion and constraint checking to prevent invalid workflow executions.","intents":["Prevent invalid parameters from being passed to n8n workflows","Provide Claude with validation feedback for parameter correction","Enforce data type and constraint requirements before workflow execution"],"best_for":["Teams running production n8n workflows that need strict input validation","Builders creating AI agents that need to learn from validation failures","Systems where invalid workflow inputs cause expensive or irreversible operations"],"limitations":["Validation rules are inferred from n8n workflow definitions — custom validation logic not supported","Complex constraints (cross-field dependencies, conditional requirements) may not be expressible in schema","Type coercion may silently convert invalid inputs (e.g., string to number) causing unexpected behavior","No support for async validation (e.g., checking if email exists in database)"],"requires":["n8n workflow input variable definitions with type information","JSON schema validation library","Parameter schema inference from n8n workflows"],"input_types":["MCP tool invocation parameters","Workflow input schemas","Validation rules"],"output_types":["Validation success/failure responses","Detailed validation error messages","Type coercion results"],"categories":["safety-moderation","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm-n8n-mcp__cap_9","uri":"capability://safety.moderation.rate.limiting.and.quota.management","name":"rate limiting and quota management","description":"Implements rate limiting on MCP tool invocations to prevent overwhelming n8n with requests, supporting per-client, per-workflow, and global rate limits. Tracks quota usage and provides feedback to MCP clients about remaining capacity.","intents":["Prevent AI agents from overwhelming n8n with rapid workflow invocations","Manage n8n resource consumption in multi-tenant scenarios","Provide quota visibility to AI agents for intelligent request scheduling"],"best_for":["Multi-tenant AI agent platforms running on shared n8n instances","Teams with rate-limited n8n deployments (e.g., cloud instances with API quotas)","Builders creating fair-use policies for AI-driven workflow access"],"limitations":["Rate limiting is enforced at MCP server layer — n8n's native rate limits not integrated","Quota tracking is in-memory — lost on server restart without persistence","No support for time-window based quotas (e.g., 100 requests per hour) — only token bucket","Rate limit configuration requires server restart — no runtime updates"],"requires":["Rate limiting configuration (limits per client, workflow, global)","Quota tracking mechanism (in-memory or external store)","Client identification for per-client rate limits"],"input_types":["MCP tool invocation requests","Rate limit configuration","Quota tracking state"],"output_types":["Rate limit enforcement (accept/reject decisions)","Quota usage metadata (remaining requests, reset time)","Rate limit error responses"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":46,"verified":false,"data_access_risk":"high","permissions":["n8n instance (self-hosted or cloud) with API access enabled","MCP client implementation (Claude Desktop, custom MCP host, etc.)","Node.js 16+ runtime for MCP server process","n8n API credentials (API key or webhook authentication)","n8n workflows with explicitly defined input variables","MCP client capable of passing structured JSON parameters","n8n API access with workflow execution permissions","n8n instance with credential management enabled","API key or OAuth token with credential access permissions","Secure credential storage (environment variables or secrets manager)"],"failure_modes":["Requires n8n instance to be running and accessible from MCP server host","Workflow discovery is static at server startup — runtime workflow creation not reflected without restart","No built-in authentication caching — each MCP call must authenticate with n8n credentials","Parameter mapping assumes n8n workflow input variables are pre-defined — dynamic variable creation not supported","No type coercion — parameter types must match workflow expectations or execution fails","Large parameter payloads (>1MB) may exceed MCP message size limits","Workflow execution is synchronous — long-running workflows block MCP response until completion","Credential types supported are limited to n8n's native credential system — custom credential types not supported","No credential rotation automation — manual credential updates required in n8n","Credentials are transmitted over HTTP if n8n instance is not HTTPS-enabled","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.6522944463099357,"quality":0.3,"ecosystem":0.68,"match_graph":0.25,"freshness":0.52,"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":"2026-05-24T12:16:23.902Z","last_scraped_at":"2026-05-03T14:04:47.472Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":91294,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=n8n-mcp","compare_url":"https://unfragile.ai/compare?artifact=n8n-mcp"}},"signature":"6icGVM7Vo04SrPn4djfEBFT7nQGfESRfWQ76Ln6QzQDuMolag4dkKlvKd6ajZO/Yho/74W0LNotvoq2cNUHNDA==","signedAt":"2026-06-21T18:49:42.458Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/n8n-mcp","artifact":"https://unfragile.ai/n8n-mcp","verify":"https://unfragile.ai/api/v1/verify?slug=n8n-mcp","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"}}