Hi Product Hunters π!
Thrilled to announce that we're open-sourcing our alerting CLI tool, Keep (https://github.com/keephq/keep) πΌ.
Designed by developers for developers, Keep streamlines and simplifies alerting, making it a first-class citizen within the development process.
Think of Keep as Prometheus Alertmanager but for all observability tools, with a simple and intuitive (GitHub actions-like) syntax.
We believe that alerting has historically been neglected in existing monitoring platforms, leading to subpar alerting practices. With Keep, we aim to change that and fight the alert-fatigue problem πͺ.
Although it's still in its early stages, we would love to get your feedback on our project.
Keep provides the following key features:
π Declarative alerting that can be easily managed and versioned in your version control and service repository.
π¨ Alerts from multiple data sources for added context and insights.
π Freedom from vendor lock-in, making it easier to switch to a different observability tool if needed.
Some of the features we plan to add in the future include:
- Integration with CI/CD processes to simplify alerts maintenance and testing.
- Scoring system to assess the alert's urgency and provide relevant information.
- Slack integration to keep track of alerts over time.
- More providers, conditions, and other enhancements.
We invite you to give Keep a try (https://github.com/keephq/keep) and share your thoughts with us. Your feedback will help us make Keep the best it can be.
@oe_zheng precisely the way you described it! We aim to be a helpful resource when fighting alert fatigue π¨ a lot more to come so stay tuned.
Thanks for your support!
@nadavwiz Thanks Wiz! and yes, of course you can customize it, take a look at the attached screenshot from Slack, where we used the "blocks" to create a custom alert.
Checkout this https://github.com/keephq/keep/b... example to see how can you achieve that using Keep and Slack's https://api.slack.com/block-kit documentation for their blocks.
Appreciate your support!
Alert fatigue is TOO real!
But getting too fancy with selective monitoring can be catastrophic - how do you guys make sure the mission critical stuff doesn't get dropped on the floor with the scoring system?
Keep