How it works Use cases Pricing Docs Swagger FAQ
Try it Join early access
One endpoint • Root cause + fix • Machine-readable JSON
Returns: root cause · severity · confidence (0–1) · category · next actions.

Explain any error instantly — via API.

Paste a stack trace, CI log, failed API response, or test failure. Get a clear root cause, severity, confidence, and a suggested fix. Built for pipelines and products — not chats.

Demo mode by default — runs locally in your browser. Live API endpoint ships next.

Returns: root cause • severity • confidence • category

This page runs in demo mode by default. A single live API endpoint is coming next.

Try a sample error View request/response
Works with JS, Python, Java, .NET CI/CD + webhook failures Redacts secrets by default

Start here (60 seconds)

  1. Open the “Try it” playground (top-right on this page).
  2. Leave it in Demo mode → click Explain error
    No API call. Runs locally in your browser.
  3. Click Use live API → confirm the API Base URL → click Explain error again
    Calls POST /v1/explain-error and returns structured trust signals.
Demo mode = local placeholder output
Live API mode = real classification, confidence, evidence & action signal

How it works

Three steps. No dashboards required.

1) Send the error

POST raw logs / stack trace / response body to /v1/explain-error.

2) We extract signal

Normalise noise, detect runtime/framework, map to known failure patterns.

3) You get structured output

Root cause, fix, severity, confidence, category, next actions — in JSON.

Try the demo (local, no API call)

Demo mode runs entirely in your browser (placeholder logic).
Live mode calls /v1/explain-error on your backend.

Demo mode
Suggest API shape Join early access
Ready. Tip: set your API base URL to call a real backend.
—
Category—
Severity—
Confidence—
Action—
{}

Where teams embed it

This becomes “debugging intelligence” your systems can call automatically.

CI/CD failures

Attach explanations to failed GitHub Actions / GitLab jobs automatically.

Support deflection

Turn scary errors into user-safe explanations + next steps for your support team.

Observability

Enrich alerts in Slack/PagerDuty with probable cause + fix on call.

Docs (MVP version)

This section is the API contract + copy/paste examples. Use it when calling the API from Postman, curl, scripts, or CI/CD. For interactive testing, use the Try it playground (top-right).

GET /health GET /v1/examples (sample requests you can paste straight into POST) GET /v1/schema (self-documenting response shape) POST /v1/explain-error POST /v1/enrich

Tip: You can open GET endpoints directly in your browser by visiting: <API_BASE>/v1/examples and <API_BASE>/v1/schema.

POST /v1/explain-error Content-Type: application/json { "text": "…raw error/log/stack trace…", "stack": "…optional stack trace…" }
{ "classification": "network/connection_refused", "confidence": 0.78, "confidenceRationale": "Matched connection refusal markers (ECONNREFUSED/connection refused).", "severity": "high", "evidence": [ { "type": "keyword_match", "value": "ECONNREFUSED", "weight": 0.45 } ], "actionSignal": "review", "explanation": "Target service refused the connection (service down, wrong host/port, or firewall).", "recommendedNextStep": "Verify service health, endpoint/port, and network ACL/security group rules." }
GET /v1/schema returns the current response shape so clients can integrate confidently. Use it as the "source of truth" for field names and types.

Uses the same API Base URL as the Try playground. If empty, it defaults to https://explain-error-api.onrender.com.

{}
curl -X POST https://explain-error-api.onrender.com/v1/explain-error \ -H "Content-Type: application/json" \ -d '{ "text": "Error: connect ECONNREFUSED 10.0.0.5:5432" }'

Accepted request fields: text (preferred), rawError, error, message, and optional stack.

Pricing (simple + dev-friendly)

Early pricing for developers exploring error explanation in real workflows.

Starter

$49 / month
For solo devs & small projects.
  • 2,000 explanations / month
  • Basic redaction
  • Email support
Try demo
Most popular

Pro

$149 / month
For SaaS teams and CI pipelines.
  • 15,000 explanations / month
  • Framework detection + categories
  • Webhook/CI templates
Join early access

Team

$399 / month
For higher volume + multiple services.
  • 50,000 explanations / month
  • Priority support
  • Basic SLA
Request Team access

FAQ

Is this just “copy/paste into ChatGPT”?
No — the MVP is designed to return structured JSON with category, severity, confidence and next actions, so it can be used in CI, alerts and tooling. Over time, your moat is a pattern library + heuristics.
How do you handle secrets in logs?
The MVP should redact common secrets (tokens, API keys, JWTs, auth headers) before analysis and never store raw logs by default.
What’s the MVP scope?
Static MVP: error → classification → confidence + severity → triage.
Next: test against real errors, then define the /v1/explain-error contract.
Can I white-label or offer an on-prem option?
Not in the initial MVP, the current focus is on a cloud-hosted model to validate the core signals and workflows.
White-label and on-prem options may be explored later if there’s clear, repeated demand.

Early access

This is a static demo designed to validate the structure, signals, and workflows before any live endpoint exists.

The focus right now is on:

Follow progress on GitHub Suggest an improvement Star the repo to stay in the loop.

Privacy note: errors pasted into the demo stay in your browser. Future behaviour will be documented explicitly before anything ships.

© ExplainError API • Built for dev teams
Docs Swagger Pricing Try Early access Feedback / Issues