# Use with agents

Information about the telemetry libraries can be shared with different agents using the workleap-telemetry agent skill or the Context7 MCP server.

# Install agent skill

Open a terminal and install the workleap-telemetry agent skill by running the following command:

npx skills add https://github.com/workleap/wl-telemetry --skill workleap-telemetry

The skills.sh CLI will prompt you to choose whether to install the skill globally or within a project. We recommend installing it locally so it is available for code review tools such as Claude Code code review.

# Try it 🚀

Once the skill is installed, start an agent and ask it to setup a project:

I'm setting up telemetry in a new React + TypeScript application. Set up Workleap telemetry end-to-end using the documented APIs and patterns.

# Setup Context7 MCP server

Context7 is an MCP server that provides AI assistants with real-time, version-specific programming documentation and code examples. Instead of relying only on a model's training data, it allows AI tools to fetch up-to-date information directly from authoritative sources (such as the code repository or the official documentation) at request time. This helps reduce hallucinations and errors caused by outdated APIs, especially in coding environments such as IDEs or editor integrations.

# VS Code

For a quick setup, install the Context7 MCP server directly from the VS Code Marketplace:

# Configuration file

Alternatively, you can configure the MCP server manually by updating the VS Code MCP configuration file:

"mcp": {
    "servers": {
        "context7": {
            "type": "http",
            "url": "https://mcp.context7.com/mcp",
            "headers": {
                "CONTEXT7_API_KEY": "YOUR_OPTIONAL_API_KEY"
            }
        }
    }
}

"mcp": {
    "servers": {
        "context7": {
            "type": "stdio",
            "command": "npx",
            "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_OPTIONAL_API_KEY"]
        }
    }
}

For more details about supported clients and configurations, refer to the Context7 clients documentation.

# Verify the connection

Open the Extensions view in VS Code and locate the context7 MCP server. Click the gear icon to access its settings and confirm that the server is running:

# Claude Code

To install Context7 with Claude Code, open a terminal and execute one of the following command:

claude mcp add --header "CONTEXT7_API_KEY: YOUR_OPTIONAL_API_KEY" --transport http context7 https://mcp.context7.com/mcp

claude mcp add context7 -- pnpmx -y @upstash/context7-mcp --api-key YOUR_OPTIONAL_API_KEY

For more details and alternative installation options, refer to the Context7 Claude Code documentation.

# Verify the connection

Open Claude Code and execute the /mcp command:

Then, select the Context7 MCP server to view its details, including connection status, authentication, and available tools:

Alternatively, you can verify the connection from a terminal by executing the claude mcp list command. The context7 MCP server should be listed as connected.

# Try it 🚀

Once the MCP server is set up, ask the agent to use Context7 to perform tasks such as generating boilerplate code:

I'm setting up telemetry in a new React + TypeScript application. Use context7 "/workleap/wl-telemetry" library  to set up Workleap telemetry end-to-end using the documented APIs and patterns.