# Datadog MCP for AI Agents AI Agent Connect > Datadog MCP lets you manage your observability stack through natural conversation. Instead of clicking through complex dashboards, you can query metrics, search logs, and manage monitors using your AI client. It gives you a direct line to your infrastructure health, making it easy to triage incidents, check SLO compliance, and keep your alerts organized without leaving your workspace. ## Overview - **Category:** loved-by-devs - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_pCs8gaBRSNau7exJae9C6MPyEaXc4N3mFANNYVEO/ai-agent-connect - **Tags:** full-stack-monitoring, infrastructure-metrics, log-analysis, incident-management, cloud-monitoring, alerting ## Description Datadog lets you connect your observability stack to your AI agent so you can manage everything through natural conversation. When a production outage hits, the last thing you want to do is navigate a labyrinth of tabs to find the right dashboard. This Connector changes that by letting your AI agent act as a direct interface for your Datadog account. You can ask questions about your system's health, and your agent will pull the data you need instantly. It handles the heavy lifting of navigating your monitors and logs so you can focus on fixing the root cause. Whether you're trying to figure out why a specific service is lagging or you need to check if your error budgets are holding up, this connection puts all that data into a single conversational flow. It's a huge time saver for anyone who lives in the telemetry data but hates the friction of manual searching. You can find this and thousands of other connections in the Vinkius catalog, where they're managed and hosted for you. It's about getting the right information without the manual overhead of clicking through every single widget. ## Tools ### check_datadog_status Verify your connection to the Datadog API is active. This helps you ensure your agent can see your data without any connectivity issues. ### create_event Create a custom event in your Datadog account with specific tags and priority levels. This is useful for marking milestones in your telemetry. ### get_dashboard Fetch the full details and layout for a specific dashboard. Use this to understand how a dashboard is structured without opening it. ### get_incident Pull all the details, including status and timeline, for a specific incident. It gives you a clear history of what happened. ### get_monitor Retrieve the specific configuration and details for a single monitor. This helps you see exactly what triggers a specific alert. ### list_dashboards Get a complete list of all dashboards in your account. Use this to browse your available monitoring views quickly. ### list_events See a list of recent events recorded in your platform. This helps you track high-level activity across your infrastructure. ### list_hosts View all reporting hosts with their associated metadata and tags. This is great for auditing your fleet's current state. ### list_incidents See a list of all currently active incidents. Use this to get a quick overview of what's currently broken. ### list_metrics Browse the metrics available in your Datadog environment. This helps you identify which data points you can query. ### list_monitors Get a list of all monitors configured in your account. Use this to audit your alerting coverage across your stack. ### list_slos See all your Service Level Objectives and their current statuses. This helps you track your performance against your targets. ### mute_monitor Silence a specific monitor to stop it from sending alerts. Use this to keep your notifications clean during maintenance. ### query_metrics Run a custom query to get specific time-series data. This lets you pull exact numbers for your specific needs. ### search_logs Search through your indexed logs using Datadog's query syntax. This is the fastest way to find errors across your stack. ### search_monitors Find specific monitors based on names or tags. This helps you locate specific alerts in a crowded environment. ## Prompt Examples **Prompt:** ``` Show me the active incidents and tell me which one is the most severe. ``` **Response:** ``` I found 3 active incidents currently running: | Severity | Status | Responder | Timeline | | :--- | :--- | :--- | :--- | | **P1** | Investigating | @dev_ops_lead | 15 mins ago | | P2 | Acknowledged | @sre_team | 45 mins ago | | P3 | Open | Unassigned | 1 hour ago | The **P1 incident** regarding 'High Latency on Checkout' is the most severe. It was triggered 12 minutes ago. ``` **Prompt:** ``` Search the logs for any 'TimeoutException' in the production environment from the last 30 minutes. ``` **Response:** ``` Searching logs with query `status:error env:production message:*TimeoutException*`... Found 42 occurrences in the last 30 minutes. **Top Sources:** - api-gateway: 28 hits - payment-service: 10 hits - auth-service: 4 hits Would you like me to pull the full stack trace for the most recent occurrence? ``` **Prompt:** ``` What's the current status of our API Availability SLO? ``` **Response:** ``` Here is the current status for your Service Level Objectives: **API Availability (99.9%)** - **Current Status:** 99.92% - **Error Budget Remaining:** 14.2% - **Burn Rate:** 0.4x (Healthy) **Checkout Latency P95 < 800ms** - **Current Status:** 99.1% - **Error Budget Remaining:** 2.1% - **Burn Rate:** 1.8x (Warning: Budget will be exhausted in 4 days at current rate). ``` ## Capabilities ### Query time-series metrics Get specific data points from your infrastructure using standard Datadog syntax. ### Search log events Find specific errors or events across all your indexed log sources. ### Manage monitor alerts List, search, and mute monitors to keep your notification volume under control. ### Inspect dashboard layouts Retrieve full details on dashboards, including widgets and template variables. ### Track active incidents Get a bird's-eye view of ongoing issues, including severity and responder info. ### Review SLO compliance Check your Service Level Objectives against targets and error budgets. ### Audit host inventory See a list of all reporting hosts along with their metadata and agent versions. ## Use Cases ### Triage a production outage An SRE asks the agent to find all active incidents and then pulls the logs for the most severe one to find the error using search_logs. ### Check SLO status during a release A platform engineer asks the agent if the API availability SLO is still healthy after a new deployment using list_slos. ### Mute noisy alerts during maintenance A DevOps engineer asks the agent to mute all monitors related to the database during a scheduled update using mute_monitor. ### Inventory audit A team lead asks the agent to list all hosts running an outdated agent version to plan a rollout using list_hosts. ## Benefits - Stop hunting for logs: Use search_logs to find errors across your entire stack instantly instead of clicking through different sources and tabs. - Triage incidents faster: Use list_incidents and get_incident to see exactly who's responding and what the timeline looks like in seconds. - Control your alerts: Use list_monitors and mute_monitor to stop noisy notifications during maintenance without having to log into the UI. - Track your budgets: Use list_slos to see exactly how much error budget you have left and which services are at risk of breaching targets. - Audit your fleet: Use list_hosts to get a clear picture of every host's agent version and metadata in one simple list. - Query custom data: Use query_metrics to pull specific time-series data using standard syntax without building a new dashboard first. ## How It Works The bottom line is you get a conversational interface for your entire Datadog observability stack. 1. Subscribe to the Datadog MCP and provide your API key and site URL. 2. Connect your preferred AI client to the Vinkius platform. 3. Ask your agent to perform tasks like searching logs or checking monitor statuses. ## Frequently Asked Questions **Can the Datadog MCP help me triage outages faster?** Yes, it lets you pull incident details and search logs through your AI client instantly, so you don't have to switch tabs to see what's broken. **How do I use the Datadog MCP to manage noisy alerts?** You can ask your agent to list your monitors and mute the ones that are firing during maintenance, helping you keep your notifications clean. **Can I check my SLO budgets with the Datadog MCP?** Yes, your agent can pull your Service Level Objectives and tell you exactly how much error budget remains for your key services. **Does the Datadog MCP support custom metric queries?** It does. You can ask your agent to run specific queries using Datadog's standard syntax to get the exact numbers you need. **Can I see my host inventory using the Datadog MCP?** Yes, your agent can list all reporting hosts along with their tags and agent versions to help you audit your infrastructure. **Is the Datadog MCP good for searching logs?** It's excellent for that. You can search across all indexed log sources using natural language commands to find specific errors or events. **Can I silence noisy monitors during scheduled maintenance?** Yes. The `mute_monitor` action silences a specific monitor by its ID, suppressing all alert notifications. This is ideal during deployment windows or planned maintenance. Use `search_monitors` to find the monitor by name or tag first, then mute it by ID. **Does Datadog require two credentials to connect?** Yes. You need your **API Key** (found in Organization Settings > API Keys) and your **Base URL**, which depends on your Datadog site region: `https://api.datadoghq.com` for US1, `https://api.datadoghq.eu` for EU, or `https://api.us3.datadoghq.com` for US3. The API Key is sent via the `DD-API-KEY` header. **Can I run time-series metric queries with custom time ranges?** Yes. The `query_metrics` tool accepts a Datadog metric query string (e.g., `avg:system.cpu.user{host:web-01}`), a start epoch timestamp, and an end epoch timestamp. It returns the time-series data points for that metric across the specified window.