Flaky test detection and quarantine
Detect flaky tests across any language, test runner, or CI provider, then quarantine them automatically while still collecting failure stack traces.
Trunk
Freemium
Trunk è una piattaforma per l’affidabilità CI che rileva test instabili, isola i fallimenti e gestisce le code di merge nei workflow GitHub, con piani gratuiti, a consumo ed enterprise.
Cos'è Trunk?
Trunk is a CI reliability platform for teams that want to keep builds green, reduce flaky test noise, and move pull requests through merge workflows faster. The product combines flaky test detection and quarantine with merge queue tooling so teams can manage test instability and merge bottlenecks in one system.
The platform is built for GitHub-based workflows and supports any language, test runner, or CI provider for flaky test handling. On the merge side, it supports batching, parallel queues, webhook and Slack notifications, and build-system-aware workflows for monorepos using tools such as Bazel, Nx, and Turborepo.
Cosa può fare Trunk?
Failure analysis and summaries
Analyze failures with test summaries in pull requests, duplicate-failure detection, and AI-generated summaries of what went wrong.
Ticketing and workflow integrations
Route failures into Linear or Jira and use webhook support to keep tickets and notifications in sync with test status.
Merge queue protection
Protect main branches with merge queue behavior that validates each PR against the latest main before merging.
Batching and parallel queues
Group PRs with intelligent batching and parallel queues to reduce redundant CI work on unrelated changes.
Queue automation and control
Use a public REST API and GitHub-based queue controls to submit, cancel, prioritize, and monitor PRs in the queue.
Casi d'uso
“Reduce flaky test noise”
Use Trunk to identify flaky tests, quarantine them automatically, and keep CI signal usable while the team continues shipping.
“Protect the main branch”
Use merge queue protection, batching, and parallel queues to keep main green and reduce manual coordination when many PRs land at once.
“Keep reviewers and contributors updated”
Use webhook support, Slack notifications, and PR comments to keep engineers informed about test status and queue progress without switching tools.
“Automate merge operations”
Use the queue controls, REST API, and GitHub integration to manage priority, cancellation, and status updates from existing workflow surfaces.
Domande frequenti
How does Trunk count users?
Trunk defines users as non-bot users who have made a commit to a private repo with Trunk enabled in the last 30 days. Contributor counts are shown on the Settings page under Billing.
What happens if I exceed the free tier limits?
Trunk uses soft limits, so you can keep using it at any scale, but you will see an upgrade banner if you exceed the free tier limits. If usage is far above the plan, Trunk may reach out about upgrading.
Can I cancel at any time?
Yes. You can cancel your Trunk account at any time.
Can I try Trunk without a credit card?
Yes. You can start using Trunk without a credit card, up to the free tier usage limits.
Does Trunk offer a free trial?
Yes. Trunk offers a free tier, and the team can also set up Team or Enterprise trials with real-time technical support and guidance.
Informazioni rapide
- Category
- CI reliability platform
- Primary use
- Flaky test detection, quarantine, and merge queue management
- Platform focus
- GitHub repositories
- Pricing model
- Free tier plus usage-based paid plans
- Enterprise options
- SSO, on-premise deployment, admin tools, custom billing
- Website
- trunk.io
Analisi del traffico di Trunk
I dati sul traffico sono solo a scopo di riferimento.
- Domain Rating
- 59
Alternative a Trunk
Typebot
typebot.io
Typebot è una piattaforma no-code per creare chatbot e moduli conversazionali, incorporarli nei siti web e collegarli a WhatsApp, webhook e servizi di automazione. Offre piani gratuito, a pagamento ed Enterprise.
Gatling
gatling.io
Gatling è una piattaforma di load testing per creare, eseguire, analizzare e automatizzare test delle prestazioni con flussi code-first, low-code e no-code.