Integrations Overview

Copilot Collections integrates with 11 external services via the Model Context Protocol (MCP). These integrations bring Jira, Figma, documentation, browser automation, structured reasoning, document reading, and cloud provider APIs directly into your Copilot sessions — enabling the end-to-end product engineering workflow.

What is MCP?

The Model Context Protocol allows VS Code Copilot agents to call external tools as part of their workflow. Each MCP server exposes specific capabilities (search, navigate, execute) that agents can use to gather information or perform actions.

Configured Servers

Server	Type	Purpose	Used By
Atlassian	HTTP	Jira & Confluence integration	BA, Architect, CR, CE, E2E, SE
Figma	HTTP	Design extraction and verification	BA, Architect, SE, UI Reviewer, CR, E2E
Context7	stdio	Library documentation search	Architect, SE, CR, UI Reviewer, E2E, Copilot Eng., DevOps
Playwright	stdio	Interactive browser automation and UI debugging	SE, E2E
Sequential Thinking	stdio	Structured reasoning for complex problems	BA, Architect, SE, CR, E2E, UI Reviewer, Copilot Eng., Orchestrator, DevOps
PDF Reader	stdio	PDF document extraction	BA, CE, Architect
AWS API	stdio	AWS infrastructure automation and resource management	DevOps
AWS Documentation	stdio	AWS service documentation and reference	DevOps
GCP Gcloud	stdio	Google Cloud operations and management	DevOps
GCP Observability	stdio	Google Cloud monitoring and observability	DevOps
GCP Storage	stdio	Google Cloud Storage integration	DevOps

Configuration

All servers are configured in .vscode/mcp.json:

{
  "servers": {
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"],
      "type": "stdio"
    },
    "context7": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp@latest"]
    },
    "sequential-thinking": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-sequential-thinking"],
      "type": "stdio"
    },
    "figma": {
      "url": "https://mcp.figma.com/mcp",
      "type": "http"
    },
    "atlassian": {
      "url": "https://mcp.atlassian.com/v1/mcp",
      "type": "http"
    },
    "pdf-reader": {
      "command": "npx",
      "args": ["@sylphx/pdf-reader-mcp"],
      "type": "stdio"
    },
    "awslabs.aws-api-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.aws-api-mcp-server@latest"]
    },
    "awslabs.aws-documentation-mcp-server": {
      "command": "uvx",
      "args": ["awslabs.aws-documentation-mcp-server@latest"],
      "env": {
        "FASTMCP_LOG_LEVEL": "ERROR",
        "AWS_DOCUMENTATION_PARTITION": "aws"
      }
    },
    "gcp-gcloud": {
      "command": "npx",
      "args": ["-y", "@google-cloud/gcloud-mcp"]
    },
    "gcp-observability": {
      "command": "npx",
      "args": ["-y", "@google-cloud/observability-mcp"]
    },
    "gcp-storage": {
      "command": "npx",
      "args": ["-y", "@google-cloud/storage-mcp"]
    }
  }
}

For Figma-backed UI verification, this MCP template is necessary but not sufficient on its own. EXPECTED comes from Figma MCP, while ACTUAL is collected by tsh-ui-capture-worker through Playwright CLI artifacts against a running app and an exact user-confirmed full dev server URL.

Server Types

• stdio — Runs locally via npx or uvx. The MCP server starts as a child process.
• HTTP — Connects to a remote API endpoint. Requires authentication (handled by the service).