mirror of
https://github.com/czlonkowski/n8n-mcp.git
synced 2026-02-06 21:43:07 +00:00
fix: comprehensive error handling and node-level properties validation (fixes #26)
Root cause: AI agents were placing error handling properties inside `parameters` instead of at node level Major changes: - Enhanced workflow validator to check for ALL node-level properties (expanded from 6 to 11) - Added validation for onError property values and deprecation warnings for continueOnFail - Updated all examples to use modern error handling (onError instead of continueOnFail) - Added comprehensive node-level properties documentation in tools_documentation - Enhanced MCP tool documentation for n8n_create_workflow and n8n_update_partial_workflow - Added test script demonstrating correct node-level property usage Node-level properties now validated: - credentials, disabled, notes, notesInFlow, executeOnce - onError, retryOnFail, maxTries, waitBetweenTries, alwaysOutputData - continueOnFail (deprecated) Validation improvements: - Detects misplaced properties and provides clear fix examples - Shows complete node structure when properties are incorrectly placed - Type validation for all node-level boolean and string properties - Smart error messages with correct placement guidance 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
czlonkowski
@@ -33,7 +33,13 @@ export class TaskTemplates {
|
||||
configuration: {
|
||||
method: 'GET',
|
||||
url: '',
|
||||
authentication: 'none'
|
||||
authentication: 'none',
|
||||
// Default error handling for API calls
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 3,
|
||||
waitBetweenTries: 1000,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -52,6 +58,11 @@ export class TaskTemplates {
|
||||
property: 'sendHeaders',
|
||||
description: 'Add custom headers if needed',
|
||||
when: 'API requires specific headers'
|
||||
},
|
||||
{
|
||||
property: 'alwaysOutputData',
|
||||
description: 'Set to true to capture error responses',
|
||||
when: 'Need to debug API errors'
|
||||
}
|
||||
]
|
||||
},
|
||||
@@ -66,7 +77,13 @@ export class TaskTemplates {
|
||||
sendBody: true,
|
||||
contentType: 'json',
|
||||
specifyBody: 'json',
|
||||
jsonBody: ''
|
||||
jsonBody: '',
|
||||
// POST requests might modify data, so be careful with retries
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 2,
|
||||
waitBetweenTries: 1000,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -84,11 +101,17 @@ export class TaskTemplates {
|
||||
{
|
||||
property: 'authentication',
|
||||
description: 'Add authentication if required'
|
||||
},
|
||||
{
|
||||
property: 'onError',
|
||||
description: 'Set to "continueRegularOutput" for non-critical operations',
|
||||
when: 'Failure should not stop the workflow'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'Make sure jsonBody contains valid JSON',
|
||||
'Content-Type header is automatically set to application/json'
|
||||
'Content-Type header is automatically set to application/json',
|
||||
'Be careful with retries on non-idempotent operations'
|
||||
]
|
||||
},
|
||||
|
||||
@@ -102,6 +125,12 @@ export class TaskTemplates {
|
||||
authentication: 'genericCredentialType',
|
||||
genericAuthType: 'headerAuth',
|
||||
sendHeaders: true,
|
||||
// Authentication calls should handle auth failures gracefully
|
||||
onError: 'continueErrorOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 3,
|
||||
waitBetweenTries: 2000,
|
||||
alwaysOutputData: true,
|
||||
headerParameters: {
|
||||
parameters: [
|
||||
{
|
||||
@@ -144,7 +173,10 @@ export class TaskTemplates {
|
||||
httpMethod: 'POST',
|
||||
path: 'webhook',
|
||||
responseMode: 'lastNode',
|
||||
responseData: 'allEntries'
|
||||
responseData: 'allEntries',
|
||||
// Webhooks should always respond, even on error
|
||||
onError: 'continueRegularOutput',
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -178,7 +210,10 @@ export class TaskTemplates {
|
||||
path: 'webhook',
|
||||
responseMode: 'responseNode',
|
||||
responseData: 'firstEntryJson',
|
||||
responseCode: 200
|
||||
responseCode: 200,
|
||||
// Ensure webhook always sends response
|
||||
onError: 'continueRegularOutput',
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -199,7 +234,12 @@ export class TaskTemplates {
|
||||
nodeType: 'nodes-base.postgres',
|
||||
configuration: {
|
||||
operation: 'executeQuery',
|
||||
query: ''
|
||||
query: '',
|
||||
// Database reads can continue on error
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 3,
|
||||
waitBetweenTries: 1000
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -229,7 +269,12 @@ export class TaskTemplates {
|
||||
operation: 'insert',
|
||||
table: '',
|
||||
columns: '',
|
||||
returnFields: '*'
|
||||
returnFields: '*',
|
||||
// Database writes should stop on error by default
|
||||
onError: 'stopWorkflow',
|
||||
retryOnFail: true,
|
||||
maxTries: 2,
|
||||
waitBetweenTries: 1000
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -265,7 +310,13 @@ export class TaskTemplates {
|
||||
content: ''
|
||||
}
|
||||
]
|
||||
}
|
||||
},
|
||||
// AI calls should handle rate limits and API errors
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 3,
|
||||
waitBetweenTries: 5000,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -393,7 +444,12 @@ return results;`
|
||||
resource: 'message',
|
||||
operation: 'post',
|
||||
channel: '',
|
||||
text: ''
|
||||
text: '',
|
||||
// Messaging can continue on error
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 2,
|
||||
waitBetweenTries: 2000
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -427,7 +483,13 @@ return results;`
|
||||
fromEmail: '',
|
||||
toEmail: '',
|
||||
subject: '',
|
||||
text: ''
|
||||
text: '',
|
||||
// Email sending should retry on transient failures
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 3,
|
||||
waitBetweenTries: 3000,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
@@ -562,6 +624,255 @@ return results;`
|
||||
'Test each tool individually before combining',
|
||||
'Set N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true for community nodes'
|
||||
]
|
||||
},
|
||||
|
||||
// Error Handling Templates
|
||||
'api_call_with_retry': {
|
||||
task: 'api_call_with_retry',
|
||||
description: 'Resilient API call with automatic retry on failure',
|
||||
nodeType: 'nodes-base.httpRequest',
|
||||
configuration: {
|
||||
method: 'GET',
|
||||
url: '',
|
||||
// Retry configuration for transient failures
|
||||
retryOnFail: true,
|
||||
maxTries: 5,
|
||||
waitBetweenTries: 2000,
|
||||
// Always capture response for debugging
|
||||
alwaysOutputData: true,
|
||||
// Add request tracking
|
||||
sendHeaders: true,
|
||||
headerParameters: {
|
||||
parameters: [
|
||||
{
|
||||
name: 'X-Request-ID',
|
||||
value: '={{ $workflow.id }}-{{ $itemIndex }}'
|
||||
}
|
||||
]
|
||||
}
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
property: 'url',
|
||||
description: 'The API endpoint to call',
|
||||
example: 'https://api.example.com/resource/{{ $json.id }}'
|
||||
}
|
||||
],
|
||||
optionalEnhancements: [
|
||||
{
|
||||
property: 'authentication',
|
||||
description: 'Add API authentication'
|
||||
},
|
||||
{
|
||||
property: 'onError',
|
||||
description: 'Change to "stopWorkflow" for critical API calls',
|
||||
when: 'This is a critical API call that must succeed'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'Retries help with rate limits and transient network issues',
|
||||
'waitBetweenTries prevents hammering the API',
|
||||
'alwaysOutputData captures error responses for debugging',
|
||||
'Consider exponential backoff for production use'
|
||||
]
|
||||
},
|
||||
|
||||
'fault_tolerant_processing': {
|
||||
task: 'fault_tolerant_processing',
|
||||
description: 'Data processing that continues despite individual item failures',
|
||||
nodeType: 'nodes-base.code',
|
||||
configuration: {
|
||||
language: 'javaScript',
|
||||
jsCode: `// Process items with error handling
|
||||
const results = [];
|
||||
|
||||
for (const item of items) {
|
||||
try {
|
||||
// Your processing logic here
|
||||
const processed = {
|
||||
...item.json,
|
||||
processed: true,
|
||||
timestamp: new Date().toISOString()
|
||||
};
|
||||
|
||||
results.push({ json: processed });
|
||||
} catch (error) {
|
||||
// Log error but continue processing
|
||||
console.error('Processing failed for item:', item.json.id, error);
|
||||
|
||||
// Add error item to results
|
||||
results.push({
|
||||
json: {
|
||||
...item.json,
|
||||
error: error.message,
|
||||
processed: false
|
||||
}
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
return results;`,
|
||||
// Continue workflow even if code fails entirely
|
||||
onError: 'continueRegularOutput',
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
property: 'Processing logic',
|
||||
description: 'Replace the comment with your data transformation logic'
|
||||
}
|
||||
],
|
||||
optionalEnhancements: [
|
||||
{
|
||||
property: 'Error notification',
|
||||
description: 'Add IF node after to handle error items separately'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'Individual item failures won\'t stop processing of other items',
|
||||
'Error items are marked and can be handled separately',
|
||||
'continueOnFail ensures workflow continues even on total failure'
|
||||
]
|
||||
},
|
||||
|
||||
'webhook_with_error_handling': {
|
||||
task: 'webhook_with_error_handling',
|
||||
description: 'Webhook that gracefully handles processing errors',
|
||||
nodeType: 'nodes-base.webhook',
|
||||
configuration: {
|
||||
httpMethod: 'POST',
|
||||
path: 'resilient-webhook',
|
||||
responseMode: 'responseNode',
|
||||
responseData: 'firstEntryJson',
|
||||
// Always continue to ensure response is sent
|
||||
onError: 'continueRegularOutput',
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
property: 'path',
|
||||
description: 'Unique webhook path',
|
||||
example: 'order-processor'
|
||||
},
|
||||
{
|
||||
property: 'Respond to Webhook node',
|
||||
description: 'Add node to send appropriate success/error responses'
|
||||
}
|
||||
],
|
||||
optionalEnhancements: [
|
||||
{
|
||||
property: 'Validation',
|
||||
description: 'Add IF node to validate webhook payload'
|
||||
},
|
||||
{
|
||||
property: 'Error logging',
|
||||
description: 'Add error handler node for failed requests'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'onError: continueRegularOutput ensures webhook always sends a response',
|
||||
'Use Respond to Webhook node to send appropriate status codes',
|
||||
'Log errors but don\'t expose internal errors to webhook callers',
|
||||
'Consider rate limiting for public webhooks'
|
||||
]
|
||||
},
|
||||
|
||||
// Modern Error Handling Patterns
|
||||
'modern_error_handling_patterns': {
|
||||
task: 'modern_error_handling_patterns',
|
||||
description: 'Examples of modern error handling using onError property',
|
||||
nodeType: 'nodes-base.httpRequest',
|
||||
configuration: {
|
||||
method: 'GET',
|
||||
url: '',
|
||||
// Modern error handling approach
|
||||
onError: 'continueRegularOutput', // Options: continueRegularOutput, continueErrorOutput, stopWorkflow
|
||||
retryOnFail: true,
|
||||
maxTries: 3,
|
||||
waitBetweenTries: 2000,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
property: 'url',
|
||||
description: 'The API endpoint'
|
||||
},
|
||||
{
|
||||
property: 'onError',
|
||||
description: 'Choose error handling strategy',
|
||||
example: 'continueRegularOutput'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'onError replaces the deprecated continueOnFail property',
|
||||
'continueRegularOutput: Continue with normal output on error',
|
||||
'continueErrorOutput: Route errors to error output for special handling',
|
||||
'stopWorkflow: Stop the entire workflow on error',
|
||||
'Combine with retryOnFail for resilient workflows'
|
||||
]
|
||||
},
|
||||
|
||||
'database_transaction_safety': {
|
||||
task: 'database_transaction_safety',
|
||||
description: 'Database operations with proper error handling',
|
||||
nodeType: 'nodes-base.postgres',
|
||||
configuration: {
|
||||
operation: 'executeQuery',
|
||||
query: 'BEGIN; INSERT INTO orders ...; COMMIT;',
|
||||
// For transactions, don\'t retry automatically
|
||||
onError: 'continueErrorOutput',
|
||||
retryOnFail: false,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
property: 'query',
|
||||
description: 'Your SQL query or transaction'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'Transactions should not be retried automatically',
|
||||
'Use continueErrorOutput to handle errors separately',
|
||||
'Consider implementing compensating transactions',
|
||||
'Always log transaction failures for audit'
|
||||
]
|
||||
},
|
||||
|
||||
'ai_rate_limit_handling': {
|
||||
task: 'ai_rate_limit_handling',
|
||||
description: 'AI API calls with rate limit handling',
|
||||
nodeType: 'nodes-base.openAi',
|
||||
configuration: {
|
||||
resource: 'chat',
|
||||
operation: 'message',
|
||||
modelId: 'gpt-4',
|
||||
messages: {
|
||||
values: [
|
||||
{
|
||||
role: 'user',
|
||||
content: ''
|
||||
}
|
||||
]
|
||||
},
|
||||
// Handle rate limits with exponential backoff
|
||||
onError: 'continueRegularOutput',
|
||||
retryOnFail: true,
|
||||
maxTries: 5,
|
||||
waitBetweenTries: 5000,
|
||||
alwaysOutputData: true
|
||||
},
|
||||
userMustProvide: [
|
||||
{
|
||||
property: 'messages.values[0].content',
|
||||
description: 'The prompt for the AI'
|
||||
}
|
||||
],
|
||||
notes: [
|
||||
'AI APIs often have rate limits',
|
||||
'Longer wait times help avoid hitting limits',
|
||||
'Consider implementing exponential backoff in Code node',
|
||||
'Monitor usage to stay within quotas'
|
||||
]
|
||||
}
|
||||
};
|
||||
|
||||
@@ -588,6 +899,13 @@ return results;`
|
||||
return this.templates[task];
|
||||
}
|
||||
|
||||
/**
|
||||
* Get a specific task template (alias for getTaskTemplate)
|
||||
*/
|
||||
static getTemplate(task: string): TaskTemplate | undefined {
|
||||
return this.getTaskTemplate(task);
|
||||
}
|
||||
|
||||
/**
|
||||
* Search for tasks by keyword
|
||||
*/
|
||||
@@ -607,13 +925,14 @@ return results;`
|
||||
*/
|
||||
static getTaskCategories(): Record<string, string[]> {
|
||||
return {
|
||||
'HTTP/API': ['get_api_data', 'post_json_request', 'call_api_with_auth'],
|
||||
'Webhooks': ['receive_webhook', 'webhook_with_response'],
|
||||
'Database': ['query_postgres', 'insert_postgres_data'],
|
||||
'AI/LangChain': ['chat_with_ai', 'ai_agent_workflow', 'multi_tool_ai_agent'],
|
||||
'Data Processing': ['transform_data', 'filter_data'],
|
||||
'HTTP/API': ['get_api_data', 'post_json_request', 'call_api_with_auth', 'api_call_with_retry'],
|
||||
'Webhooks': ['receive_webhook', 'webhook_with_response', 'webhook_with_error_handling'],
|
||||
'Database': ['query_postgres', 'insert_postgres_data', 'database_transaction_safety'],
|
||||
'AI/LangChain': ['chat_with_ai', 'ai_agent_workflow', 'multi_tool_ai_agent', 'ai_rate_limit_handling'],
|
||||
'Data Processing': ['transform_data', 'filter_data', 'fault_tolerant_processing'],
|
||||
'Communication': ['send_slack_message', 'send_email'],
|
||||
'AI Tool Usage': ['use_google_sheets_as_tool', 'use_slack_as_tool', 'multi_tool_ai_agent']
|
||||
'AI Tool Usage': ['use_google_sheets_as_tool', 'use_slack_as_tool', 'multi_tool_ai_agent'],
|
||||
'Error Handling': ['modern_error_handling_patterns', 'api_call_with_retry', 'fault_tolerant_processing', 'webhook_with_error_handling', 'database_transaction_safety', 'ai_rate_limit_handling']
|
||||
};
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user