Cross-platform terminal and web system monitor for CPU, memory, disk, network, processes and containers, with a REST API and exports.
- Stars
- 12,420
- Forks
- 1,533
- Open issues
- 664
- Last commit
- 22 days ago
What is it?
Keep is an open-source alert and incident management platform for operations teams. It collects alerts from many monitoring, logging and incident tools into a single interface, where they can be deduplicated, enriched, filtered and correlated before anyone has to act on them.
Workflows are defined as declarative YAML files with triggers, steps and actions. A workflow can run when an alert arrives, on a schedule or manually, fetch context from other systems, and then open a ticket, send a message or call another provider. The project lists a broad set of integrations, including Prometheus, Datadog, Grafana, PagerDuty, Opsgenie, Slack, Jira and Kubernetes, with some integrations supporting bi-directional sync.
Keep also offers AI-backed enrichment and correlation through providers such as OpenAI, Anthropic and Ollama. It can be run locally, with Docker Compose, on Kubernetes, on AWS ECS or on OpenShift, and it supports SSO, SAML, OIDC and LDAP authentication with role-based access control.
Key features
- 01Aggregates alerts from many monitoring and incident tools into one interface
- 02Deduplicates, correlates, filters and enriches alerts
- 03Declarative YAML workflows triggered by alerts, schedules or manually
- 04Bi-directional integrations with ticketing, chat and on-call tools
- 05AI-powered enrichment and summarization via providers such as OpenAI and Ollama
- 06SSO, SAML, OIDC and LDAP authentication with RBAC
Use cases
- Reducing alert noise by deduplicating and correlating alerts from several monitoring systems
- Automatically creating Jira or GitHub tickets for critical alerts that match certain filters
- Sending alerts to Slack or on-call tools with added context from other systems
- Running self-hosted alert management on Kubernetes or in air-gapped environments
Quick start
- 1
Try the hosted platform in your browser
Things to know
- The license is listed as NOASSERTION, so the license terms could not be confirmed from the metadata.
Topics
FAQ
- What is keep?
- Open-source AIOps platform that consolidates alerts from monitoring tools, deduplicates and correlates them, and automates responses via workflows. Keep is an open-source alert and incident management platform for operations teams. It collects alerts from many monitoring, logging and incident tools into a single interface, where they can be deduplicated, enriched, filtered and correlated before anyone has to act on them.
- Is keep free to use? What license does it have?
- keep is open source under the Other license. Check the license terms before using it in commercial products.
- Is keep still maintained?
- Actively maintained. The last commit to keep was on Sep 28, 2026.
- What can I use keep for?
- Reducing alert noise by deduplicating and correlating alerts from several monitoring systems Automatically creating Jira or GitHub tickets for critical alerts that match certain filters Sending alerts to Slack or on-call tools with added context from other systems Running self-hosted alert management on Kubernetes or in air-gapped environments
- How do I install keep?
- Try the hosted platform in your browser
- Where can I get keep?
- The source code and installation instructions are on GitHub: https://github.com/keephq/keep
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Lightweight self-hosted server monitoring with a web dashboard, Docker container stats, historical data and alerts.
Open-source observability and APM tool that uses eBPF to collect metrics, logs, traces and profiles, with AI-assisted root cause analysis.
Open-source distributed monitoring system for tracking the performance and availability of networks, servers, VMs, applications and cloud services.
Checkmk is an open-source IT monitoring system for infrastructure and applications across physical, virtual, container and cloud environments.
A self-hosted AI SRE agent that learns normal log patterns, auto-raises incidents for new anomalies, and routes alerts to Slack, Teams, Telegram and on-call tools.