Build on ViviScape Work.
Three ways in, one platform. Every path signs in with the same identity, respects the same permissions, and writes to the same activity log — pick the one that fits the job.
How do I build on ViviScape Work?
Use the REST API when your own system talks to Work, the MCP server when an AI agent does the work, and the vs CLI when a script or a person at a terminal does. Starter gets read-only REST; full read and write access starts at Pro, and the MCP server is part of Premium.
REST API
Raw HTTPS to api.viviscape.io. Authenticate once, then send a bearer token to any endpoint under /api/v1, from any language.
Best for: production integrations — syncing customers from your CRM, pushing time into payroll, building dashboards or a customer portal.
Explore the APIMCP server
Connect Claude, ChatGPT, OpenAI Codex or any MCP client — through the hosted server at mcp.viviscape.io, or the viviscape-mcp package installed locally.
Best for: AI agents that pick up tasks, log time, chat in Spaces and leave notes, each under its own sign-in and permissions.
Set up MCPCLI (vs)
ViviScape.CLI — a global .NET tool with a gh auth login-style browser sign-in and --format json on every command.
Best for: scripts, CI pipelines, bulk imports, scheduled reports and quick one-off queries from a terminal.
Get the CLIViviScape Work developer questions
Does ViviScape Work have an API?
Yes. ViviScape Work has a REST API covering customers, projects, tasks, time entries and invoices. The Starter plan gets read-only access; full read and write access is included from the Pro plan upward.
What is the ViviScape CLI?
The ViviScape CLI is a command-line tool, invoked as vs, for scripting work against a ViviScape Work workspace — bulk imports, scheduled reports, CI hooks and one-off queries. It requires a paid plan.
Can AI agents connect to ViviScape Work?
Yes, through the ViviScape MCP server, which exposes the workspace over the Model Context Protocol to any MCP-capable assistant or agent. Use the hosted server at mcp.viviscape.io for ChatGPT and Claude.ai, or install the viviscape-mcp package for Claude Desktop, Claude Code and OpenAI Codex. An agent signs in either as you or as itself, with its own permissions, and every action lands in the activity log under that identity.
Can I import historical time entries from another system?
Yes. Historical hours can be imported through the REST API or the vs CLI. Most teams start logging new work in ViviScape Work on day one, let the old system finish the current billing cycle, and backfill history afterwards.
Pick a path. Ship today.
REST for production systems. MCP for AI agents. CLI for scripts and ops. Same identity. Same audit trail.