Introduction
BugBoard turns runtime errors into cards on a kanban board. Instead of a separate error inbox you triage somewhere else, every reported error becomes a task you can prioritise, assign, comment on, and move to done — in the same board your team already works from.
There is exactly one public ingestion endpoint — POST /api/v1/tasks — and
everything else is built on top of it.
Two ways to report
You can integrate BugBoard in two ways. They hit the same endpoint and produce the same cards.
| Method | Best for | What you write |
|---|---|---|
| HTTP API (recommended) | Any language or platform, full control, no dependency | A signed (or bearer) POST request |
| SDK | JavaScript/TypeScript and PHP apps | A few lines of setup + named helpers |
We recommend the API as the source of truth. It works everywhere, has no moving parts to keep up to date, and the SDKs are a thin, ergonomic wrapper around it. Reach for an SDK when you want the queue, retries, and shutdown flushing handled for you.
How it fits together
your app ──► POST /api/v1/tasks ──► BugBoard ──► card on your board
(API key + payload) dedupes, triage, assign,
routes to a comment, resolve
board column
- A report carries a severity (
critical | major | moderate | minor) and a priority (low | medium | high). Together they decide how the card looks and where it lands. - Repeat reports are deduplicated into a single card with an occurrence count, so a spike of the same error doesn't flood your board. They don't have to be identical — reports sharing a title group together, and each distinct payload is kept beneath the card as a variant you can browse. See Deduplication.
Where to go next
- Quickstart — send your first card in a few minutes.
- Choosing a key type — publishable vs. secret keys.
- API Reference — the complete, canonical contract.
- SDKs — JavaScript/TypeScript and PHP guides.