What your AI apps can do with Datadog
- Search logs, metrics and traces
- Review monitors, dashboards and incidents
Datadog MCP tools
31 tools, taken from the publisher's documentation on 29 October 2026. The server decides what it offers and can add or rename tools at any time; after you connect, Toolstile shows the current list.
| Tool | What it does |
|---|---|
search_datadog_logs | Searches logs |
analyze_datadog_logs | Runs SQL-based statistics over logs |
get_datadog_metric | Queries historical or real-time metric data |
search_datadog_metrics | Lists metrics and reads their tags and metadata |
get_datadog_metric_context | Lists metrics and reads their tags and metadata |
get_datadog_trace | Reads a full APM trace by ID |
search_datadog_spans | Finds spans and computes counts, percentiles and timeseries |
aggregate_spans | Finds spans and computes counts, percentiles and timeseries |
search_datadog_monitors | Reads monitor status, thresholds and conditions |
search_datadog_incidents | Lists incidents or reads one |
get_datadog_incident | Lists incidents or reads one |
search_datadog_events | Searches alerts, deploys and infrastructure change events |
search_datadog_hosts | Lists hosts and host info |
search_datadog_entities | Reads Software Catalog ownership and dependencies |
search_datadog_rum_events | Searches and aggregates RUM events |
aggregate_rum_events | Searches and aggregates RUM events |
search_datadog_dashboards | Lists dashboards or reads one |
get_datadog_dashboard | Lists dashboards or reads one |
search_datadog_notebooks | Lists notebooks or reads one |
get_datadog_notebook | Lists notebooks or reads one |
create_datadog_notebook | Creates or edits a notebook |
edit_datadog_notebook | Creates or edits a notebook |
delete_datadog_notebook | Soft-deletes a notebook (recoverable for 30 days) |
upsert_datadog_dashboard | Creates or updates a dashboard |
delete_datadog_dashboard | Permanently deletes a dashboard (dashboards toolset) |
create_datadog_monitor | Creates a monitor as a draft that sends no notifications (alerting toolset) |
search_datadog_error_tracking_issues | Searches Error Tracking issues (error-tracking toolset) |
update_datadog_error_tracking_issue | Changes an issue's state or assignee (error-tracking toolset) |
update_datadog_feature_flag_environment | Changes a flag's configuration in one environment (feature-flags toolset) |
sync_datadog_feature_flag_allocations | Replaces all allocations in an environment (feature-flags toolset) |
execute_code | Runs agent-written JavaScript in a sandbox with your access (code-exec toolset) |
Suggested access for each AI app
The documentation this list comes from doesn't mark Datadog's tools as read-only or destructive. Read each tool's description and decide per app before allowing it.
Coding agent in Claude Code or Cursor. Allow the read tools so the agent can look things up without stopping, and require approval for tools that change shared data.
Chat or research app in Claude Desktop. Allow reading and searching. Allow writes only where you want the app to act for you.
Unattended agent, such as a scheduled Codex job. No one is watching the session, so keep it narrow: allow what the job needs, require approval for changes and block the rest.
Blocked tools are removed from the tool list an app sees, so it won't try them. Because rules apply per app, the same Datadog connection can be broad in one app and read-only in another, and the credential stays in Toolstile instead of each app's config file.
How it works with Toolstile
- Connect once. Add Datadog to a workspace (browser sign-in). Toolstile keeps the credential, so no agent config holds it.
- Choose access per app. Give Claude Code every Datadog tool, give Cursor only the read tools, or keep Datadog away from an app entirely.
- Share approved accounts. Teammates use the connection without copying tokens, and you can remove access in one place.
- Review activity. See which app called which tool and when, so you can tighten access with evidence.
Add the Datadog MCP server to an AI app
Through Toolstile, every AI app uses one gateway entry, whichever servers you connect. In Claude Code:
claude mcp add --transport http toolstile 'https://mcp.toolstile.com/mcp'Then connect Datadog in your workspace and choose which of its tools that app may use. Client setup covers Codex, Cursor and Claude Desktop.
To connect directly instead, add the Datadog endpoint to each app separately. In Claude Code:
claude mcp add --transport http datadog 'https://mcp.datadoghq.com/api/unstable/mcp-server/mcp'In Cursor's mcp.json:
{
"mcpServers": {
"datadog": {
"url": "https://mcp.datadoghq.com/api/unstable/mcp-server/mcp"
}
}
}Some servers accept only clients they have approved, or need a token header. Check the Datadog MCP documentation for what your client needs.
Before you connect
A Datadog account on a supported site. Your role's permissions apply.
Server details
| Detail | Value |
|---|---|
| Publisher | Datadog |
| Endpoint | https://mcp.datadoghq.com/api/unstable/mcp-server/mcp |
| Authentication | Browser sign-in |
| Categories | Development |
| Tool list | 31 tools, observed 29 October 2026 (docs) |