Watch review progress in real time

During reviews, the MCP server connects to the Optibot backend via WebSocket and emits real-time progress notifications using MCP logging messages. Your MCP client will receive updates as the review progresses through these steps:

1. started — Review request accepted

2. analyzing_patch — Parsing and analyzing the diff

3. tool_call — Running analysis tools (with tool name and query details)

4. generating_review — Generating the final review

5. completed — Review finished

CI/CD Integration

The Optibot MCP server can be used in CI/CD pipelines for automated code reviews. Use the OPTIBOT_API_KEY environment variable for headless (non-interactive) authentication — no browser login is needed.

Setting Up

1. Generate an API key from the Optibot dashboard or using the create_api_key tool.

2. Store it as a secret in your CI provider (e.g., GitHub Actions secret, GitLab CI variable).

3. Pass it as OPTIBOT_API_KEY in your pipeline environment.

GitHub Actions Example

- name: Optibot Review
  env:
    OPTIBOT_API_KEY: ${{ secrets.OPTIBOT_API_KEY }}
  run: npx @optimalai/optibot-mcp

Other CI Providers

Set the OPTIBOT_API_KEY environment variable in your pipeline configuration. The MCP server will automatically use it for authentication without requiring interactive login.

Environment Variables

VariableDescriptionOPTIBOT_API_KEYYour API token (required for automated use)OPTIBOT_API_URLCustom backend URL (defaults to https://agents.getoptimal.ai)