Embrace MCP Server adds web support and new tools for spans, logs, and exceptions

Previously, the Embrace MCP Server was only available on our supported mobile platforms of iOS, Android, React Native, and Unity. Now, the Embrace MCP Server is available for web apps! Web apps can query data from all applicable tools, including our app, exception, log, session, span, and network tools!

Here’s a short demo video showing how you can easily explore the health and traffic for a web app with the Embrace MCP Server.

We’ve added span data to the Embrace MCP Server, and it currently has the following tools:

• list_root_spans: List all root spans in an app with performance metrics and outcome rates
• get_root_span_distribution: Get distribution by various dimensions for a specific root span

You can start exploring high-level data across your spans with the following prompts:

• What’s the current health of my app’s performance spans?
• Which spans have the worst p90 latency?
• Which spans have the highest error rate?
• Has anything changed since yesterday in the performance of my app’s spans?

If you discover a root span that you want to investigate further, the Embrace MCP Server can analyze this data across many dimensions, including:

• Failure rates filtered by or grouped by standard attributes (e.g., country, OS version, app version)
• Key latency metrics (e.g., p50, p90) as well as duration distributions
• Impact across users, sessions, and specific app versions

Here’s a short demo video showing you can search for issues and latency insights in your spans.

We’ve added log data to the Embrace MCP Server, and it currently has the following tools:

• list_logs: List top log messages grouped by aggregated message pattern
• get_log_details: Get detailed information about a specific log group including message template and token analysis
• get_log_distribution: Get distribution of a log group across a dimension (OS version, device model, country, etc.) compared against session baselines

You can start exploring high-level data across your logs with the following prompts:

• Show me the error logs for this app.
• Show me any changes in logs since the last app release.
• Show me which logs are affecting the most users.
• Are there any notable commonalities for my user cart checkout log?

You can then drill down into specific logs for more details or to look at distributions.

Here’s a short demo video showing you can investigate log distributions across dimensions like browser and app version.

We’ve added exception data to the Embrace MCP Server, and it currently has the following tools:

• list_exceptions: List top exceptions ranked by frequency and user impact
• get_exception_details: Get detailed information about a specific exception group, including counts, affected users, and history

You can start exploring high-level data across your exceptions with the following example prompts:

• Are there any new exceptions in the website today?
• Which exceptions have the highest severity score?
• Did we introduce any new exceptions in the latest release?
• Tell me how I can fix or mitigate my top exception given my codebase here: @[path_to_project_dir]

Quick note: The Embrace MCP Server includes the severity score for JavaScript exceptions in web apps, giving you even more context to prioritize issues directly in your AI workflows.

Here’s a short demo video highlighting how you can query for exceptions, rank them by their severity score, and then explore possible code fixes.

We’ve expanded our crash support to include a new crash tool:

• get_crash_distribution: Get distribution of a crash group across a dimension (OS version, device model, country, etc.) compared against session baselines

In addition to ranking crashes by frequency and pulling stack traces for crash analysis, you can now investigate the distribution of specific crashes across key dimensions like app version and OS version.

Wrapping up

We’ve added many new ways to explore your mobile (and now web!) data with the Embrace MCP Server, and we look forward to your feedback as you incorporate these real user insights into your AI tools and workflows.

If you’d like to try out the Embrace MCP Server, head to our docs to learn how to configure it, get more information on supported use cases, and start integrating Embrace observability data into your development and AI workflows.