n8n Failed Execution Doctor

Unofficial community tool. This project is not affiliated with, endorsed by, or sponsored by n8n GmbH.

Diagnose a failed n8n execution without connecting to the customer's live n8n instance.

Provide two JSON objects:

• the exported n8n workflow;
• the failed execution JSON.

The Actor returns the failing node, node type, normalized error category, sanitized error message, and a concrete next diagnostic step.

Why this is different

This is not another generic workflow linter. It focuses on incidents that already happened: failed runs, timeouts, authentication failures, rate limits, invalid API payloads, expression/item-linking errors, and network failures.

It is deterministic: no external LLM call is required for the core diagnosis, which keeps latency and platform cost low.

Free Lite n8n workflow

Want to run the same core diagnosis entirely inside n8n? Download the Lite workflow JSON or inspect examples/n8n-failed-execution-doctor-lite.json.

The Lite workflow uses only two core n8n nodes (Execute Workflow Trigger + Code), requires no external API key or LLM, and returns the failing node, root-cause category, sanitized error message, and a focused next diagnostic step for one incident item.

The exact JSON in this repository has been imported successfully into an isolated n8n 2.33.7 runtime. It is inactive by default.

Pro workflow — batch diagnosis and incident reports

If you need a self-contained n8n workflow rather than the Actor, n8n Failed Execution Doctor Pro v0.3.0 adds:

• batch diagnosis for multiple failed-execution items in one run;
• deterministic severity and retry-safety heuristics;
• workflow risk hints;
• ready-to-copy Markdown incident reports;
• sample input/output and a commercial internal-use license.

It requires no external LLM, API key, database, or live n8n login. The Pro workflow has been imported and executed end-to-end on n8n 2.33.7.

Pro details and purchase: https://n8n-doctor.167-233-67-162.sslip.io/?src=github

Reliability Toolkit bundle — $29

If you also want to catch workflow risks before deployment, the Reliability Toolkit Pro includes Incident Doctor Pro plus a zero-dependency preflight scanner, Markdown/JSON reports, broken-connection and webhook checks, domain/credential inventory, and a ready-to-use GitHub CI template.

Toolkit details: https://n8n-doctor.167-233-67-162.sslip.io/?src=github

Lite vs Pro

Capability	Free Lite	Pro v0.3.0
Failing/last node detection	✓	✓
Root-cause classification	✓	✓
Sanitized error evidence	✓	✓
Focused next diagnostic step	✓	✓
Batch incident diagnosis	—	✓
Severity heuristic	—	✓
Retry-safety guidance	—	✓
Workflow risk hints	—	✓
Markdown incident report	—	✓
Sample input/output package	—	✓

Privacy boundary

The Actor does not need n8n credentials and does not log in to a live n8n instance. Common secret-like strings in error messages are redacted and returned messages are truncated. Users should still remove unnecessary sensitive execution data before submitting JSON.

Categories detected

• authentication: 401/403, authorization, credential and token failures;
• rate_limit: HTTP 429 and rate-limit responses;
• timeout: timeout, ETIMEDOUT and ECONNABORTED;
• network: connection, DNS, socket and reachability failures;
• invalid_input: HTTP 400/422, validation and malformed request failures;
• expression: undefined fields, expression and item-linking problems;
• execution_error: structured failures that do not match a more specific class.

Output

Each run writes one diagnosis to the default dataset and to the DIAGNOSIS key-value record. A no-error input returns NO_STRUCTURED_ERROR_FOUND rather than inventing a root cause.

Apify deployment

The code supports Apify pay-per-event through the incident-diagnosed event. Configure that event in Apify Console and set ACTOR_CHARGE_EVENT_NAME=incident-diagnosed for the production Actor.

Local PPE validation has been completed with Apify's test mode and a single charge event per completed diagnosis.