Skip to content
Aerial view of the Petronas Twin Towers in Kuala Lumpur at night

Error capture · Built in Malaysia

Every blip,
traced in any language.

Bliptrace is one central backend for error telemetry from every app you ship. Your app sends the error and gets a 202 in milliseconds. We scrub it, group it and put it in front of your team.

Start freeFollow an error
Accepted before any storage write
202
Official SDKs across 8 runtimes
21
PII stored. Scrubbed on arrival
0
Issue per bug, not per occurrence
1

The hot path

Accept-fast, flush-later

Five cheap, in-memory steps, then 202. Storage happens after the response, in batches. A slow database never slows the app you are monitoring, and a full buffer sheds with 429 instead of blocking.

  1. 01

    Validate

    Envelope checked against the versioned JSON Schema.

  2. 02

    Resolve

    Project comes from the ingest key — never from the payload.

  3. 03

    Scrub

    PII stripped before anything is stored or returned.

  4. 04

    Fingerprint

    Per-runtime normalisers hash the top in-app frames.

  5. 05

    Sample

    Per-node quota and sampling, in-process.

  6. 202 Accepted

    Your app moves on. Bliptrace batch-flushes occurrences to ClickHouse and issues to Postgres.

What you get

From raw telemetry to a triaged inbox

Close-up of source code on a monitor

Grouped issue inbox

Per-runtime normalisers understand PHP, minified JavaScript, Python and Go stack traces. The same bug becomes one issue with a count, not a thousand alerts.

project_idexception.typetop in-app framesculprit

PII scrubbed on arrival

On the client and the server, before anything is stored, returned or forwarded. Ready for government projects.

"email": "[redacted]"

Lifecycle triage

Open, acknowledge, resolve, mute or reopen. Assign issues and get alerted when a resolved one comes back.

openackresolvedmuted
Server racks with network cabling in a data centre

Two stores, each doing one job

Postgres holds issues and everything mutable. ClickHouse holds every occurrence, append-only, with TTL. No Redis, no Kafka.

Ask your errors through MCP

An MCP server lets your AI assistant read issues and occurrences directly. Writes such as resolve and mute go through the same audited path as the dashboard.

> what broke after release 2.4?

City lights across the Earth seen from orbit at night

Runtime-agnostic

One backend for every runtime you ship

Drop in an SDK, or POST the error envelope from any HTTP client. No lock-in.

  • php
  • laravel
  • symfony
  • cakephp
  • codeigniter
  • yii2
  • drupal
  • wordpress
  • js
  • express
  • nextjs
  • python
  • django
  • flask
  • fastapi
  • ruby
  • rails
  • go
  • rust
  • java
  • dotnet
  • + any HTTP client
A development team working together at laptops around a table

For the whole team

Stop hearing about bugs from your users

Roles that fit
Owner, admin, member and viewer, scoped per project.
Alerts where you already are
New-issue and regression alerts by email, Slack, Telegram or webhook.
Trends at a glance
Occurrence sparklines and analytics across projects and environments.
Kuala Lumpur skyline at dusk

Catch the next error before your users do.

Create a project, copy the ingest key, and you are live in minutes.

Start monitoring freeLog in