Cursor

Supported OS Linux Windows Mac OS

Overview

Datadog supports Cursor in two separate ways:

The Datadog Extension for Cursor helps you improve your software by providing code-level insights directly in the IDE, based on real-time observability data from:

  • Logs
  • Error Tracking
  • CI Visibility
  • App and API Protection
  • Synthetics
  • Static Analysis
  • Datadog Model Context Protocol (MCP) server

Use this when you want Datadog context directly in Cursor while coding.

The Cursor integration in Datadog ingests Cursor usage and spend telemetry (for example, cursor.daily_usage.*, cursor.usage_events.*, and cursor.spending.*) so you can monitor adoption, usage patterns, and cost in Datadog.

Use this when you want to monitor Cursor activity from Datadog dashboards, notebooks, and monitors.

Setup

Choose the setup flow based on which product you are configuring.

Installation and configuration

Installation (Cursor IDE)

From Cursor:

  1. Open the Extensions view and search for Datadog. Alternatively, find the Datadog Extension in the Open VSX registry.
  2. Click Install to download and install the extension in your IDE.
  3. If prompted that Datadog is a third-party extension, click Accept.
  4. Click Restart IDE.

Configuration (Cursor IDE)

You can find the extension settings within the IDE settings, under Datadog.

In the extension settings, select the Datadog services relevant to your project.

Installation (Datadog)

  1. In Datadog, open the Cursor integration tile.
  2. In Cursor, log in to the Cursor Admin Portal with an admin account.
  3. Go to Settings.
  4. Under Advanced, open Admin API Keys and generate a new Admin API key.
  5. Copy the generated Admin API key.
  6. In Datadog, add an account name and paste the key into the API Key field.
  7. Choose the telemetry you want to collect:
    • Enable Daily Usage Data
    • Enable Spending Data
    • Enable Usage Events
    • Enable AI Commit Metrics (Enterprise)
    • Enable AI Code Change Metrics (Enterprise)
  8. Save the configuration, then confirm metrics are arriving by searching for cursor. in Datadog Metrics Explorer.

Configuration (Datadog)

After setup, configure monitors and dashboards for the cursor.* metrics that matter to your teams (for example, usage volume or spend trends).

The AI Commit Metrics and AI Code Change Metrics options are available for Cursor Enterprise organizations.

Data Collected

Metrics

cursor.daily_usage.total_lines_added
(gauge)
Total number of lines added in a day.
Shown as unit
cursor.daily_usage.total_lines_deleted
(gauge)
Total number of lines deleted in a day.
Shown as unit
cursor.daily_usage.accepted_lines_added
(gauge)
Total number of suggested line additions accepted in a day.
Shown as unit
cursor.daily_usage.accepted_lines_deleted
(gauge)
Total number of suggested line deletions accepted in a day.
Shown as unit
cursor.daily_usage.total_applies
(gauge)
Total number of code change suggestions accepted in a day.
Shown as unit
cursor.daily_usage.total_rejects
(gauge)
Total number of code change suggestions rejected in a day.
Shown as unit
cursor.daily_usage.total_tabs_shown
(gauge)
Total number of autocomplete suggestions shown in a day.
Shown as unit
cursor.daily_usage.total_tabs_accepted
(gauge)
Total number of autocomplete suggestions accepted in a day.
Shown as unit
cursor.daily_usage.composer_requests
(gauge)
Total number of composer requests made in a day.
Shown as unit
cursor.daily_usage.bugbot_usages
(gauge)
Total number of bugbot usages in a day.
Shown as unit
cursor.daily_usage.chat_requests
(gauge)
Total number of chat requests made in a day.
Shown as request
cursor.daily_usage.agent_requests
(gauge)
Total number of agent requests made in a day.
Shown as request
cursor.daily_usage.cmdk_usages
(gauge)
Total number of inline edit prompt bar calls in a day.
Shown as unit
cursor.daily_usage.subscription_included_requests
(gauge)
Total number of subscription included requests made in a day.
Shown as request
cursor.daily_usage.api_key_requests
(gauge)
Total number of API key requests made in a day.
Shown as request
cursor.daily_usage.usage_based_requests
(gauge)
Total number of usage based requests made in a day.
Shown as request
cursor.spending.spend_cents
(gauge)
Total number of cents spent.
Shown as cent
cursor.spending.fast_premium_requests
(gauge)
Total number of fast premium requests made.
Shown as request
cursor.usage_events.input_tokens
(count)
Total number of input tokens used for a session.
Shown as unit
cursor.usage_events.output_tokens
(count)
Total number of output tokens used for a session.
Shown as unit
cursor.usage_events.cache_write_tokens
(count)
Total number of cache write tokens used for a session.
Shown as unit
cursor.usage_events.cache_read_tokens
(count)
Total number of cache read tokens used for a session.
Shown as unit
cursor.usage_events.total_cents
(count)
Total number of cents spent for a session.
Shown as cent
cursor.ai_commits.tab_lines_added
(count)
Total number of lines added using tab completion in a commit.
Shown as unit
cursor.ai_commits.tab_lines_deleted
(count)
Total number of lines deleted using tab completion in a commit.
Shown as unit
cursor.ai_commits.composer_lines_added
(count)
Total number of lines added using the Composer in a commit.
Shown as unit
cursor.ai_commits.composer_lines_deleted
(count)
Total number of lines deleted using the Composer in a commit.
Shown as unit
cursor.ai_commits.non_ai_lines_added
(count)
Total number of lines added without AI assistance in a commit.
Shown as unit
cursor.ai_commits.non_ai_lines_deleted
(count)
Total number of lines deleted without AI assistance in a commit.
Shown as unit
cursor.ai_changes.lines_added
(count)
Total number of lines added in a specific code change.
Shown as unit
cursor.ai_changes.lines_deleted
(count)
Total number of lines deleted in a specific code change.
Shown as unit

Support

Need help? Contact Datadog support.

For issues specific to the Datadog Extension for Cursor, open an issue on GitHub.

Further Reading