n8n-nodes-autoheal
v0.1.1
Published
n8n community node for AutoHeal AI — send errors for AI analysis and get fixes
Downloads
385
Maintainers
Readme
n8n-nodes-autoheal
n8n community node for AutoHeal AI — send production errors for AI analysis and get fixes automatically.
What is AutoHeal AI?
AutoHeal captures JavaScript/TypeScript errors, analyzes them with AI, and generates platform-specific fix prompts. It works with Lovable, Replit, Bolt, v0, Cursor, Windsurf, and more.
Install
In your n8n instance:
- Go to Settings > Community Nodes
- Enter
n8n-nodes-autoheal - Click Install
Or install manually:
npm install n8n-nodes-autohealSetup
- Sign up at autohealai.com
- Go to Settings > API Key and copy your
ah_...key - In n8n, add AutoHeal AI credentials with your API key
Operations
Send Error
Send an error to AutoHeal for AI analysis. Returns an error_id you can use to check status or get the fix.
Inputs:
- Error Message (required) — e.g.
TypeError: Cannot read properties of undefined - Stack Trace — full stack trace
- Source URL — where the error happened
- Additional Info — user agent, environment, etc.
Check Status
Check if an error has been analyzed yet.
Inputs:
- Error ID (required) — from the Send Error response
Returns: new, analyzing, analyzed, fix_applied, or ignored
Get Fix
Get the AI-generated fix for an analyzed error.
Inputs:
- Error ID (required)
- Wait for Analysis — poll until analysis completes (default: true)
- Max Wait — timeout in seconds (default: 60)
Returns: Full error details + fix explanation, root cause, suggested diff, fix prompt, and PR URL.
Example Workflow
[Error Trigger] → [AutoHeal: Send Error] → [Wait 30s] → [AutoHeal: Get Fix] → [Slack: Send Fix]- Your n8n workflow catches an error
- AutoHeal analyzes it with AI (~15-30 seconds)
- Get the fix explanation and suggested code changes
- Forward the fix to Slack, email, or wherever you want
Use Cases
- Catch failed webhook deliveries → send to AutoHeal → get fix in Slack
- Monitor cron job failures → auto-create GitHub issues with AI fix suggestions
- Aggregate errors from multiple services → centralized AI analysis
- Pipe Sentry/PagerDuty alerts → enrich with AI fix prompts
License
MIT — Knod, Inc.
