Give your AI agent a seat in the workspace.
The ViviScape MCP server hands Claude, ChatGPT, OpenAI Codex and any MCP client typed tools for CRM, projects, tasks, Spaces, time, notes, flows and insights. Use the hosted server with nothing to install, or run it locally next to your code.
Does ViviScape Work have an MCP server for AI agents?
Yes, two ways. ViviScape hosts an MCP server at mcp.viviscape.io for ChatGPT and Claude.ai connectors, and publishes the same server as the viviscape-mcp package for a local install in Claude Desktop, Claude Code, OpenAI Codex and other MCP clients. There are no API keys: you sign in as yourself, or an agent signs in as itself, and every action lands in the activity log under that identity.
Hosted server (no install)
For ChatGPT developer mode and Claude.ai custom connectors. In Work, create the agent and its credential under AI Workforce → Agents → Sign-in & access. Then add a connector with this URL and choose OAuth; when the ViviScape connect page opens, paste the credential and click Connect.
https://mcp.viviscape.io/mcp
Clients that can send their own headers can skip OAuth and send Authorization: Bearer vsa_… on every request. Revoking the credential in Work disconnects the client.
Local install: sign in once
Runs on Windows, macOS and Linux with Node.js 18 or later. Sign-in uses the same browser flow as the vs CLI — if you already ran vs auth login, that session is picked up automatically.
# opens ViviScape Work in your browser — click Allow access npx viviscape-mcp login # who am I, which account, role, plan npx viviscape-mcp status
Add it to your MCP client
Claude Desktop (claude_desktop_config.json) or a project .mcp.json:
{
"mcpServers": {
"viviscape": {
"command": "npx",
"args": ["-y", "viviscape-mcp"]
}
}
}
Claude Code and OpenAI Codex:
# Claude Code claude mcp add viviscape -- npx -y viviscape-mcp # OpenAI Codex — ~/.codex/config.toml [mcp_servers.viviscape] command = "npx" args = ["-y", "viviscape-mcp"]
Optional environment: VIVISCAPE_AGENT_TOKEN (an agent’s own vsa_… credential — every call then runs as that agent instead of you) ·
VIVISCAPE_AGENT_KEY (a display label for comments posted from your session) ·
VIVISCAPE_BASE_URL (defaults to https://work.viviscape.io).
“npx not found”? On macOS, use the absolute path that which npx prints as the command. On Windows, use "command": "cmd" with "args": ["/c", "npx", "-y", "viviscape-mcp"].
Tools exposed to your agent
Typed tools across these domains. Each call runs as the signed-in person or agent and is checked against that identity’s permissions — for an agent, the permissions and Spaces / projects / tasks its owner granted.
prospect_add, prospect_query, client_add, client_get, company_list, follow-ups and notes
project_add, project_update, project_tasks, project_staff, milestone_add, milestone_clone
task_add, task_update, task_complete, task_search, tasks_open, tasks_bulk_add, assignees, tags, references, plans and solutions
room_list, room_messages, room_message_add, room_files — agents and people working in the same room
timelog_add, task_log_time, personal_log_add, timeclock_punch, timeclock_history
note_add, notes_query, kb_search, kb_sop, media_center
flow_list, flow_start, flow_status — start a published flow and follow its run
hours by person / service / project, AI summary, account_info, account_users, user_lookup, enums
Built for agents: list tools page and trim their rows (fields, limit, offset), accept updated_since for delta reads, and creates accept an idempotency_key so a retried step does not write twice.
ViviScape MCP questions
Do I need to install anything to use the ViviScape MCP server?
Not for ChatGPT or Claude.ai. ViviScape hosts the server at https://mcp.viviscape.io/mcp: add it as a connector with OAuth and paste the agent's credential on the connect page. Claude Desktop, Claude Code and OpenAI Codex run the viviscape-mcp package locally through npx, which needs Node.js 18 or later.
Does the ViviScape MCP server use API keys?
No. Locally, you sign in once with npx viviscape-mcp login, the same browser flow as the vs CLI, and every tool call runs as you. An agent that should act as itself uses its own credential, which starts with vsa_ and is created in Work under AI Workforce, Agents, Sign-in & access.
What can an agent do through the MCP server?
Agents get typed tools for CRM, projects and milestones, tasks, Spaces and room chat, time logs and the time clock, notes, Second Brain, the Media Center, ViviScape Flow and insights. Each call is checked against the signed-in identity's permissions, and for an agent, the Spaces, projects and tasks its owner granted.
How do I disconnect an agent from ViviScape Work?
Revoke the agent's credential in Work. A hosted connection stops working immediately, and the agent's ID and history are kept. Rotating the credential means reconnecting the client once.
Which plan includes the ViviScape MCP server?
Connecting outside agents through the ViviScape MCP server is part of the Premium plan. AI Studio, the built-in assistant, is available on every plan.
Connect your first agent.
Set the agent up in Work, paste its credential into the connector, and it starts working under its own name. Connecting outside agents is part of the Premium plan.