# Opsgenie MCP > Opsgenie helps you automate your incident response. Connect it to any agent and manage alerts, track who is on-call, and coordinate major outages conversationally. Instead of clicking through dashboards during a crisis, simply ask your AI client to acknowledge an alert, check the current rotation schedule, or create a full incident report. ## Overview - **Category:** industry-titans - **Price:** Free - **Tags:** alerting, on-call, incident-response, sre, monitoring, workflow-automation ## Description Running an SRE team means living in a constant cycle of alerts and escalations. This MCP gives you control over that entire process directly from your preferred agent. You can talk to it like talking to a teammate: 'What's wrong with the database?' or 'Who owns this incident right now?'. It lets you handle everything from simple alert management—like adding notes and closing tickets—to coordinating major incidents with defined priority levels. Need to know who is covering the primary schedule next week? Just ask. You can list all schedules, check specific on-call personnel, or query historical data across both alerts and incidents. When you connect it via Vinkius, your agent gains immediate access to this powerful operational visibility. It turns a stressful, multi-tab manual process into simple conversation. ## Tools ### acknowledge_alert Marks a specific alert as acknowledged so that response time metrics are accurate. ### add_note Attaches textual information or investigation findings to an existing alert record. ### close_alert Closes a specific alert when the root cause has been resolved and verified. ### create_alert Generates an entirely new Opsgenie alert based on specified criteria. ### create_incident Initiates a major incident record with defined priority and scope. ### get_alert Retrieves all current details for one specified alert ID. ### get_incident Pulls the full history and status of a specific major incident. ### get_who_is_on_call Checks the current on-call schedule to identify the responsible team member or individual. ### list_alerts Gives a comprehensive list of all currently open and historical alerts in the system. ### list_incidents Lists multiple major incidents that have occurred or are currently active for review. ### list_schedules Displays all defined on-call rotation schedules and their next changes. ## Prompt Examples **Prompt:** ``` List all currently open P1 alerts. ``` **Response:** ``` I've found 3 open P1 alerts. 1. 'Database latency high' (ID: 4930), 2. 'API Gateway 5xx errors' (ID: 4931), 3. 'Payment service down' (ID: 4932). ``` **Prompt:** ``` Acknowledge alert 4930 and add a note saying 'I am investigating the connection pool limits'. ``` **Response:** ``` Alert 4930 has been acknowledged and the note has been added to the activity log. You are now the owner of this alert. ``` **Prompt:** ``` Who is currently on-call for the 'SRE-Primary' schedule? ``` **Response:** ``` The current on-call responder for 'SRE-Primary' is Sarah Miller (sarah.miller@example.com). Her shift ends at 08:00 AM UTC tomorrow. ``` ## Capabilities ### Acknowledge an alert Mark an open alert as seen so the rest of the team knows you're working on it. ### Add notes to alerts and incidents Log technical details or investigation steps directly into the activity log for full audit history. ### Close an alert or incident Formally mark a problem as resolved once troubleshooting is complete. ### Create new alerts and incidents Manually trigger the process for an outage when automated systems fail to do so. ### Check on-call status Find out who is currently assigned to handle a specific schedule or service. ### List schedules and alerts View comprehensive lists of all team rotations, active alerts, and past incidents for auditing. ## Use Cases ### The Support Ticket Triage A support agent receives a ticket about payment failures. Instead of reading the internal runbook, they ask their agent to check for active alerts. The agent runs 'list_alerts', finds an open P1 alert, and tells the agent: 'There's already an incident open; I'll add a note linking this ticket.' This saves 15 minutes of searching. ### The Critical Outage Response An API service fails at 3 AM. The on-call engineer asks their agent: 'Who is responsible for the auth gateway?' The agent runs 'get_who_is_on_call' and provides the name instantly, allowing the engineer to start communicating immediately. ### Post-Mortem Documentation The team finished an outage. Instead of copying details from three different dashboards, the Incident Manager asks their agent to 'get_incident' details and list all related alerts. The agent compiles a clean summary for the post-mortem report. ### Proactive Monitoring A junior engineer wants to check if their service is covered by an alert. They ask: 'Are there any open alerts for the payment processing queue?' The agent runs 'list_alerts' and confirms that no issues are currently flagged, giving them peace of mind. ## Benefits - Stop hunting for information. You can ask your agent to check the current rotation schedule or list all open P1 alerts instantly, eliminating manual dashboard navigation. - Maintain a perfect record of events by using 'add_note' directly through your AI client. Every action and piece of investigation data stays linked to the right alert or incident. - Accelerate response time. Instead of finding the contact email, simply ask your agent who is on-call for a given schedule via 'get_who_is_on_call', routing issues immediately. - Streamline outage recovery. After fixing something, you can use 'close_alert' and 'create_incident' to formally document the timeline, keeping audit logs clean and accurate. - Get full visibility into past events. Use 'list_alerts' or 'list_incidents' to query historical data instantly when performing a root cause analysis. ## How It Works The bottom line is that your agent talks directly to Opsgenie, retrieving and manipulating real incident data without you ever leaving your chat window. 1. Subscribe to this MCP and provide your Opsgenie API Key (GenieKey). 2. Optionally, specify the region you operate in—US or EU. 3. Start talking to it through your AI client: 'List all open P1 alerts' or 'Who is on-call for SRE?' ## Frequently Asked Questions **How do I check if an alert has already been acknowledged using Opsgenie MCP?** You can use the 'get_alert' tool to pull all details for a specific alert ID. This data includes the activity log, which will confirm who owns it and when it was last updated. **Can I list all active P1 alerts using Opsgenie MCP?** Yes, you can use 'list_alerts' to pull a comprehensive list of open issues. You simply need to phrase your query to filter for the severity level and status you care about. **Is Opsgenie MCP better than just using the native Opsgenie UI?** The key difference is context switching. Instead of leaving your IDE or chat app, your agent runs the commands for you. It brings the dashboard data directly into natural conversation. **What information does 'get_who_is_on_call' provide?** It gives you the name and contact details of the current on-call responder for a specific schedule, plus when their shift ends. This is crucial for routing issues correctly. **How do I create an incident using Opsgenie MCP?** You use the 'create_incident' tool by providing the necessary scope and priority details. The agent handles forming the record, making sure it gets logged correctly in the system.