MCP Server

1stLine by Burava provides an MCP server for teams that want to manage alert workflows from AI tools or automation.

It is especially useful for Alert Schemas work: creating schemas from payloads, updating schema configuration, and testing Schema Patterns or Transformation Template changes before rollout.

MCP endpoint

For 1stLine Cloud, use:

https://1stline.burava.com/api/mcp

If you run 1stLine with a custom base path, keep the /api/mcp route and use your own domain/base URL.

Authentication and organization context

The MCP server expects a Bearer token in either:

• Authorization
• X-Authorization

For organization-scoped tools, also include the organization context your client should work against.

• X-Org-Context: <organization_uid>

You can create and manage API Tokens in Access Control.

Use these placeholders in the examples below:

• <api_token>: your 1stLine API Token value
• <organization_uid>: the organization UID the MCP tools should use

If your client supports environment variables in MCP configuration, prefer FIRSTLINE_API_TOKEN for the token. The organization UID is not a secret, but it still must match the workspace you want the tools to operate in.

Client configuration examples

The 1stLine MCP server uses Streamable HTTP at:

https://1stline.burava.com/api/mcp

Codex

Codex stores MCP configuration in ~/.codex/config.toml. You can add the server with the Codex CLI:

export FIRSTLINE_API_TOKEN="<api_token>"
codex mcp add firstline --url https://1stline.burava.com/api/mcp --bearer-token-env-var FIRSTLINE_API_TOKEN

Then add the organization context header in ~/.codex/config.toml:

[mcp_servers.firstline]
url = "https://1stline.burava.com/api/mcp"
bearer_token_env_var = "FIRSTLINE_API_TOKEN"
http_headers = { "X-Org-Context" = "<organization_uid>" }

Verify it:

codex mcp list

OpenAI documents Codex MCP setup in the OpenAI Docs MCP guide.

Claude Code

Claude Code can add remote HTTP MCP servers from the CLI:

claude mcp add --transport http firstline https://1stline.burava.com/api/mcp \
  --header "Authorization: Bearer <api_token>" \
  --header "X-Org-Context: <organization_uid>"

For a project-scoped config, create .mcp.json in the project root:

{
  "mcpServers": {
    "firstline": {
      "type": "http",
      "url": "https://1stline.burava.com/api/mcp",
      "headers": {
        "Authorization": "Bearer <api_token>",
        "X-Org-Context": "<organization_uid>"
      }
    }
  }
}

Verify it:

claude mcp list

In Claude Code, run /mcp to confirm the server is connected. Anthropic documents this format in Connect Claude Code to tools via MCP.

Cursor

For project-specific setup, create .cursor/mcp.json in the project root. For global setup, create ~/.cursor/mcp.json.

{
  "mcpServers": {
    "firstline": {
      "url": "https://1stline.burava.com/api/mcp",
      "headers": {
        "Authorization": "Bearer ${env:FIRSTLINE_API_TOKEN}",
        "X-Org-Context": "<organization_uid>"
      }
    }
  }
}

Set the token before starting Cursor:

export FIRSTLINE_API_TOKEN="<api_token>"

Cursor documents the mcp.json format and supported transports in Model Context Protocol (MCP).

GitHub Copilot in VS Code

GitHub Copilot in VS Code uses VS Code MCP configuration. For project-specific setup, create .vscode/mcp.json in the project root. You can also use the command palette action MCP: Open User Configuration for a user-level config.

{
  "servers": {
    "firstline": {
      "type": "http",
      "url": "https://1stline.burava.com/api/mcp",
      "headers": {
        "Authorization": "Bearer ${input:firstline-api-token}",
        "X-Org-Context": "<organization_uid>"
      }
    }
  },
  "inputs": [
    {
      "type": "promptString",
      "id": "firstline-api-token",
      "description": "1stLine API Token",
      "password": true
    }
  ]
}

Open Copilot Chat, switch to Agent mode, and enable the firstline server in the tools picker. Microsoft documents this format in the VS Code MCP configuration reference.

OpenCode

OpenCode uses the mcp section in opencode.json. Use a project opencode.json, or a global config at ~/.config/opencode/opencode.json.

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "firstline": {
      "type": "remote",
      "url": "https://1stline.burava.com/api/mcp",
      "enabled": true,
      "oauth": false,
      "headers": {
        "Authorization": "Bearer {env:FIRSTLINE_API_TOKEN}",
        "X-Org-Context": "<organization_uid>"
      }
    }
  }
}

Set the token before starting OpenCode:

export FIRSTLINE_API_TOKEN="<api_token>"

Verify it:

opencode mcp list

OpenCode documents this format in MCP servers.

Antigravity

Antigravity stores custom MCP server configuration in ~/.gemini/antigravity/mcp_config.json. Open the MCP Store, click Manage MCP Servers, then View raw config.

{
  "mcpServers": {
    "firstline": {
      "serverUrl": "https://1stline.burava.com/api/mcp",
      "headers": {
        "Authorization": "Bearer <api_token>",
        "X-Org-Context": "<organization_uid>"
      }
    }
  }
}

Google documents this format in Antigravity Editor: MCP Integration.

Caution

Some MCP clients store raw configuration in local files. If your client does not support secret prompts or environment-variable substitution for HTTP headers, avoid committing that file and treat it as a local secret-bearing config.

Supported MCP tool groups

The current MCP tools cover:

• Alert Schemas
• Library Schemas
• Transformation Template testing
• Alert Instances
• Incidents
• Incident Postmortems
• Users
• Escalation Flow

For Alert Schema operations, you can list, read, create, update, delete, and test schema behavior. This includes the schema creation-token flow and pattern/template testing flows.

Library Schema tools can list, read, create, update, and delete global or organization-scoped Library Schemas, depending on the token permissions.

Incident tools can list and read Incidents. Incident Postmortem tools can read and update the postmortem document for an Incident.

User tools can list Users for the current organization context. Escalation Flow tools can resolve the flow for selected entities.

How this connects to Alert Schemas

Use MCP when your team wants a repeatable way to:

• create schemas from real producer payload examples
• test extraction/transformation changes before applying them
• keep schema changes in a workflow that can be reviewed by engineering

Then confirm the final result in Alert Schemas and Alert Instances.

Related docs

• Use MCP Server to Create Alert Schema
• Alert Schemas
• MCP Troubleshooting

Codex, Claude Code, Cursor, GitHub Copilot, VS Code, OpenCode, and Antigravity are third-party brands. Burava does not own, represent, or speak for those brands.