Add Enterprise-Ready Observability to Bright Data Web MCP via TrueFoundry

In this guide, you will learn:

• What TrueFoundry is and how it helps organizations build, deploy, and govern generative AI and ML applications.
• How TrueFoundry can add observability capabilities to the already enterprise-ready Bright Data Web MCP.
• How to configure a TrueFoundry MCP Gateway for the Web MCP and enable enterprise-grade monitoring.

Let’s dive in!

What Is TrueFoundry?

TrueFoundry is an enterprise AI platform for building, deploying, governing, and monitoring AI applications, agents, and MCP servers. It provides the infrastructure, security, and operational tooling needed to run production AI workloads across cloud, on-premises, hybrid, or air-gapped environments.

Main Features

TrueFoundry gives you the building blocks required to operate enterprise AI systems at scale, including:

• A unified gateway for managing traffic across LLMs, AI agents, and MCP servers.
• A centralized MCP server control panel for registering, discovering, securing, and connecting both third-party and custom MCP servers.
• End-to-end observability with tracing, metrics, latency analysis, and request logs for agents, models, and MCP tools.
• Enterprise security and governance through SSO, RBAC, audit logging, policy enforcement, and compliance-ready architecture.
• Infrastructure management with GPU orchestration, autoscaling, resource optimization, and model serving.
• Flexible deployment options, including public cloud, VPC, on-premises, hybrid, and air-gapped environments.
• Monitoring and alerting for performance, resource usage, throughput, latency, token consumption, and operational costs.
• Compliance-ready architecture designed to support enterprise security requirements, including SOC 2, HIPAA, and GDPR.

Building an Enterprise-Ready MCP Deployment with TrueFoundry and Bright Data Web MCP

MCP makes it easier for AI agents to discover and interact with external services (e.g., APIs and data sources). However, for a truly unstoppable AI agent, you need more than simply connecting an MCP server.

Enterprise MCP environments need to handle large-scale usage while providing reliability, security, and control. Organizations need solutions that can scale, enforce access policies, provide detailed monitoring, and offer full visibility into every interaction. Without these features, debugging issues, managing costs, and ensuring compliance becomes challenging.

This is where combining Bright Data Web MCP with TrueFoundry creates a production-ready MCP architecture!

Bright Data Web MCP: Enterprise Web Access for AI Agents

Bright Data Web MCP provides AI agents with access to more than 70 tools for web search, source discovery, browser automation, scraping, and structured data extraction. These tools help agents retrieve live web data from across the internet, including websites protected by advanced anti-bot systems.

Even in Rapid mode (5k free requests per month), Web MCP provides fundamental tools for everyday web workflows:

By enabling [Pro mode](https://github.com/brightdata/brightdata-mcp#-pricing, modes), you gain access to the full Web MCP catalog. This includes tools for structured data extraction from platforms such as Amazon, LinkedIn, Google Maps, and other popular sources. It also provides browser automation capabilities for AI-powered web workflows.

The Web MCP is engineered for enterprise workloads thanks to the infrastructure behind it. In detail, it runs on Bright Data’s global network of more than 400 million residential IPs. This supports unlimited concurrency, has 99.99% uptime, and achieves a 99.95% request success rate.

TrueFoundry MCP Gateway: Adding Enterprise Control and Observability

TrueFoundry MCP Gateway acts as a centralized layer between AI agents and MCP servers. It enables you to register, discover, secure, and manage MCP servers through a unified interface.

By placing TrueFoundry in front of the Bright Data Web MCP, you gain enterprise capabilities such as OAuth-based authentication, role-based access control, request tracing, and detailed analytics. Every MCP interaction can be monitored, helping you understand tool usage, identify issues, and optimize agent workflows. Learn more in the official docs.

Together, Bright Data Web MCP and TrueFoundry provide a complete enterprise MCP stack. Bright Data delivers scalable web intelligence capabilities, while TrueFoundry adds the governance, security, and visibility required to run MCP workloads reliably in production.

How to Set up an MCP Gateway to Bright Data Web MCP in TrueFoundry

In this guided section, you will learn how to set up a TrueFoundry MCP Gateway to connect to Bright Data Web MCP. The gateway will forward requests to the Web MCP server while adding enterprise-ready visibility, monitoring, and operational insights.

Follow the instructions below!

Prerequisites

To follow along with this section, make sure you have the following:

• A TrueFoundry account (the free trial is enough).
• A Bright Data account with an API key configured. Follow the official guide to generate your Bright Data API key.

While not strictly required, it is also helpful to:

• Be familiar with the tools exposed by Bright Data Web MCP.
• Have a basic understanding of how MCP works.

Step #1: Get Familiar with Bright Data’s Web MCP

A TrueFoundry MCP Gateway forwards requests to a remote Bright Data Web MCP server. Before configuring the gateway, it is helpful to understand the connection parameters supported by the Bright Data Web MCP.

Consider the Web MCP remote connection URL format for Pro mode:

https://mcp.brightdata.com/mcp?token=<YOUR_BRIGHT_DATA_API_KEY>&pro=1

Note that the URL contains two query parameters:

• token: Authenticates requests through your Bright Data account. Replace the <YOUR_BRIGHT_DATA_API_KEY> placeholder with your actual Bright Data API key.
• pro=1: Enables access to the full catalog of 70+ tools. Without this parameter, only the free tools available in Rapid mode are accessible (up to 5,000 requests per month).

Now, assume that you want more control. For example, you would like to expose only specific tools or tool groups. In that case, generate a custom connection URL directly from the Bright Data dashboard.

Log in to your Bright Data account and reach the “AI Gateways > MCP” page. Follow the setup wizard to configure a custom Web MCP connection URL:

Notice the additional query parameters included in the generated Streamable HTTP URL. These parameters reflect the configuration you selected.

Cool! You now know how the Bright Data Web MCP connection URL is structured and how to customize the solution for your use case.

Step #2: Configure an MCP Gateway for Web MCP in TrueFoundry

If you have not already done so, start by creating a TrueFoundry account and setting up an organization. Otherwise, simply sign in.

Once you are in the TrueFoundry dashboard, open the “MCP Gateway” section from the navigation menu:

On the “MCP Registry” page, select the “Bring your own Remote MCP Server” option. Alternatively, click “Add Server” and choose “Connect any Remote MCP Server”:

Next, fill out the “Add New MCP Server” form like this:

• Name: bright-data-web-mcp
• Description: An all-in-one MCP server that lets you search the web, discover new sources, interact with web pages, and retrieve data without getting blocked.
• URL: https://mcp.brightdata.com/mcp?token=<YOUR_BRIGHT_DATA_API_KEY>&pro=1 (or the Web MCP URL you configured in the previous step).
• Auth Data: false

Note: If you prefer not to expose your Bright Data API key in the URL, follow the documentation for setting up and using secrets in TrueFoundry.

Finally, press “Add MCP Server”. The Bright Data Web MCP server will now be configured in TrueFoundry. Well done!

Step #3: Explore the Web MCP TrueFoundry Connection

Back in the “MCP Registry” page, you should now see a card for your bright-data-web-mcp server:

If you click “Add to Client”, you will receive configurations/instructions for popular MCP clients, including Cursor, VS Code, Claude Code, Claude Web, and more.

Now, press the “bright-data-web-mcp” card to open its details page:

There, you will find the complete list of tools exposed by the Bright Data Web MCP to the TrueFoundry MCP gateway. When using Pro mode, you will see all 70+ available tools.

Pay particular attention to the URL displayed at the top of the page. In this case, if you named your MCP server bright-data-web-mcp, it should look like this:

https://gateway.truefoundry.ai/<TRUEFOUNDRY_ORG_NAME>/mcp/bright-data-web-mcp/server

This is the Streamable HTTP endpoint for connecting to the MCP Gateway. You can use this URL to connect any MCP-compatible client to your Bright Data Web MCP server via TrueFoundry. See how in the next step!

Step #4: Add the TrueFoundry Web MCP Gateway Connection to Your AI Agent

To check that the TrueFoundry MCP Gateway is correctly exposing the Bright Data Web MCP server, configure it in OpenAI Codex CLI (or your favorite MCP-compatible AI agent).

Note: You can follow the same process with any AI application that supports remote MCP servers. If you are using a different AI tool, refer to its official documentation for MCP connection instructions.

Assuming Codex CLI is already installed and configured on your system, add the bright-data-web-mcp server by running:

codex mcp add bright-data-web-mcp, url <YOUR_TRUEFOUNDRY_MCP_GATEWAY_URL>

As mentioned before, <YOUR_TRUEFOUNDRY_MCP_GATEWAY_URL> should be:

https://gateway.truefoundry.ai/<TRUEFOUNDRY_ORG_NAME>/mcp/bright-data-web-mcp/server

For direct configuration via STDIO, refer to our tutorial on how to use the Bright Data Web MCP in OpenAI Codex CLI.

The TrueFoundry MCP Gateway relies on OAuth 2.0 for authentication. After running the command, your default browser will open and display the following authorization page:

Click “Approve” to authorize Codex to access the MCP Gateway. This grants your local Codex CLI installation permission to connect to the remote Bright Data Web MCP server through the TrueFoundry gateway.

After the OAuth flow completes, your browser will display the “Authentication complete. You may close this window.” message. Meanwhile, your terminal will show output similar to the following:

Note: If you are running Codex CLI on a server without a graphical browser, copy the authorization URL printed in the terminal. Open it in a browser on another machine, complete the OAuth flow, and then return to the original terminal.

Now launch Codex CLI and run the /mcp command. You should see the bright-data-web-mcp server listed as connected through OAuth, exposing the same set of tools shown earlier in the TrueFoundry dashboard:

This is it! OpenAI Codex (or any MCP-compatible AI client) is now connected to the TrueFoundry MCP Gateway, which securely exposes the Bright Data Web MCP server.

Step #5: Test the Integration

To verify that everything is working correctly, open Codex CLI and run a prompt like this:

Using the Bright Data Web MCP, analyze how people perceive Notion across the web. Collect recent discussions from Reddit, Trustpilot, blog reviews, and other public review platforms. Identify the most common positive and negative themes, then produce a sentiment report with representative examples and actionable recommendations.

This task requires web search, source discovery, and web scraping. These capabilities are not available natively in OpenAI models available in Codex CLI, but they are exposed through the Web MCP server. That makes the above prompt a great way to test the setup. Prompts covering other web-based use cases are equally suitable.

Run the prompt, and you should see a workflow similar to this:

During the run, the AI agent performed the following steps:

1. Called the discover tool (backed by Bright Data’s Discover API) to find relevant sources for G2, Capterra, and Trustpilot reviews about Notion.
2. Used the search_engine_batch tool (built on Bright Data’s SERP API) to perform multiple Google searches for relevant queries.
3. Identified the most relevant pages, including Notion’s Trustpilot, G2, and Capterra review pages, along with several blog reviews.
4. Utilized scrape_batch together with the scrape_as_markdown tool (using Bright Data’s Web Unlocker API) to retrieve the content of each page.
5. Combined all the retrieved information into a single, final report.

Notice that the generated report is based on live, up-to-date information retrieved directly from the web. It aggregates reviews and discussions from multiple sources (Trustpilot, G2, Capterra, Reddit, and blog posts) to provide a comprehensive overview of how users perceive Notion.

This demonstrates that the TrueFoundry MCP Gateway successfully exposes the Bright Data Web MCP server. It also confirms that your AI agent can harness the Web MCP tools to complete complex web intelligence tasks. Fantastic!

Step #6: Explore the Observability Stats and Metrics

Now, return to the TrueFoundry dashboard. Open the “Metrics” page, switch to the “MCP Metrics” tab, and choose the “MCP servers” option:

There, you can monitor the following metrics in visual charts:

• Requests per minute
• Request latency
• Request failures per minute, grouped by error type
• Total request failures per minute
• Request failures breakdown
• MCP method calls breakdown
• Error breakdown

You can also filter the data by method and time range. That gives you both a real-time view of your MCP traffic and a historical snapshot of recent activity.

Then, switch to the “Tools” tab:

This section provides more granular insights into how the Bright Data Web MCP tools are being used, including:

• Tool requests per minute
• Tool request latency
• Tool request failures per minute, grouped by error type
• Total tool request failures per minute
• Request latency summary (ms)
• Request count
• Tool request failures breakdown

Finally, open the “Request Traces” page to inspect every request that passes through your MCP Gateway. For each MCP request, you can view its timestamp, execution time, input parameters, output, and other execution details.

For example, inspect a request made to the scrape_batch Web MCP tool:

These features give you complete visibility into your Bright Data Web MCP deployment. With detailed metrics, request tracing, and usage analytics, TrueFoundry adds the enterprise-grade monitoring and operational insights needed to manage Web MCP server connections at scale. Mission complete!

Conclusion

In this blog post, you learned what TrueFoundry is and how it helps bring enterprise capabilities to AI applications. Specifically, you saw why MCP servers need monitoring and observability features to become truly production-ready.

You then learned how to set up a TrueFoundry MCP Gateway for the Bright Data Web MCP. The result is an MCP server integration enriched with enterprise-grade telemetry and analytics via TrueFoundry. Combined with Bright Data’s high availability, unlimited concurrency, and 99.99% uptime, this duo creates an MCP deployment built for enterprise AI applications.

Sign up for a Bright Data account today for free and start exploring our AI-ready web data solutions!