Quick Start
Get Responder running and investigate your first alert in minutes.
Deployment
Self-host Responder on your own infrastructure.
Agents
Learn how investigation agents work and how to configure them.
Integrations
Connect Slack, Sentry, Datadog, GitHub, and 15+ other tools.
How Responder Works
Responder watches your alert sources and automatically kicks off an investigation whenever something fires. Each investigation is handled by an Agent you configure — specifying which tools to query, which AI model to use, and how to deliver results.1
Connect your tools
Link Slack, Sentry, Datadog, GitHub, and any other observability or cloud services your team uses. Responder keeps credentials encrypted per workspace.
2
Create an Agent
Configure an Agent with a trigger (Sentry project, Datadog monitor, Slack channel), the context sources to query during an investigation, and your preferred AI model.
3
Alerts trigger investigations
When an alert fires, Responder automatically runs an investigation — querying logs, traces, metrics, and code — then produces a structured report with severity, root cause, timeline, and remediation options.
4
Review results and ship fixes
View investigation reports in the dashboard or Slack. Approve an auto-generated pull request to deploy the fix, or follow the recommended remediation steps.
Key Features
Investigations
Structured reports with SEV-1/2/3 severity, root cause, evidence, and timeline — delivered to Slack or the dashboard.
Auto-Remediation
Responder can propose and open a GitHub pull request to fix the issue directly.
Issues Tracking
Issues persist across investigations so recurrences are automatically linked to their history.
Workspace Secrets
Securely inject API keys and credentials into investigation sandboxes.