generated: '2026-06-20'
method: searched
status: published
source: >-
  https://developer.remote.com/docs/introduction-to-remote-mcp and
  https://github.com/remoteoss/remote-for-ai. Remote operates an official
  hosted/remote MCP server for AI agents. Tool families captured from the MCP
  documentation; the server advertises its exact tool list to clients at
  connection time via MCP discovery.
description: >-
  Remote.com's official MCP server lets MCP-standard AI agents act on Remote
  data. Authentication is OAuth 2.0 Authorization Code with PKCE through the
  normal Remote browser sign-in with Dynamic Client Registration — no API keys,
  client secrets, or scope configuration; the agent inherits the permissions of
  the signed-in Remote account. Read tools cover the full HR/payroll surface;
  write tools are limited to employee self-service actions.
server:
  name: remote
  transport: http
  auth: oauth2
  oauth:
    flow: authorization_code_pkce
    dynamic_client_registration: true
    notes: Client registers dynamically, sends the user through Remote sign-in, and receives access scoped to that account; nothing to hardcode.
install:
  marketplace: https://github.com/remoteoss/remote-for-ai
  clients: [Claude Desktop, claude.ai, Claude Code, Cursor, Codex, Gemini CLI]
  local_toolkit:
    name: "@remoteoss/ai-agent-toolkit"
    install: npx -y @remoteoss/ai-agent-toolkit
    source: https://github.com/remoteoss/ai-agent-toolkit
    note: TypeScript toolkit (34+ tools) exposing the same surface to LangChain and MCP; backs a local/stdio server.
tool_families:
  read:
    - Employment records and employees
    - Contracts and amendments
    - Payroll, schedules, and deadlines
    - Time off and leave policies
    - Org structure
    - Legal entities and country requirements
    - Timesheets, billing, and invoices
  write:
    - Submit time off requests
    - Update personal details and contact info
    - Create/update career goals and journal notes
notes: >-
  Clients discover the concrete tool list automatically at connection time.
  Governance recommendations for each underlying operation are in
  agentic-access/remote-com-agentic-access.yml.