TL;DR. Hypertab is an AI-agent-native active table where columns don’t just hold data, they do things. Smart columns call APIs, compute formulas, perform lookups, and push to integrations for every record. Connect any MCP-compatible AI agent (Claude Code, Cursor, Windsurf) and it can build data pipelines for you.

Why we built it

Workflow tools and programmable Tables use different execution models. Event workflows make each run explicit; Hypertab keeps bounded per-record work and its results inside a Table. The right choice depends on the job shape, provider limits, and measured runtime capacity.

Hypertab brings the data and work into one Table. Records x columns can represent many repeatable pipelines, with AI agents as first-class operators via MCP.

What makes it different

Columns DO things

Every column has a kind: static (holds data), http (API call per record), formula (compute), integration (push), waterfall (try sources in order), lookup (cross-table VLOOKUP), or extract (pull a JSON field at zero cost).

Record-DAG execution

Smart columns declare dependencies via {{column}} references. Hypertab builds a dependency graph and runs each record through it in topological order, A → B → C → D. The current Free beta allows up to five concurrent record DAGs per Table, with provider and runtime controls applying additional bounds. Paid-plan concurrency is not currently available.

MCP-native

86 MCP tools out of the box. Your AI agent can create Projects and Tables, insert records, configure smart columns, run pipelines, and check progress, all from inside Claude Code or Cursor.

Designed for bounded batch work

Rate limiting, retry controls, and circuit breaking reduce the impact of expected provider failures. Cloudflare Workers serve requests and Turso stores data, while virtualized rendering limits the browser to an active record window. Actual throughput and supported size depend on record width, provider behavior, and measured account limits.

Try it

Open app. After signup, copy the generated MCP config into a compatible client to work with your authorized workspace.

{
  "mcpServers": {
    "Hypertab": {
      "url": "https://api.hypertab.ai/mcp",
      "headers": { "Authorization": "Bearer ht_sk_YOUR_KEY" }
    }
  }
}

We’re just getting started. Comparison guides, integration deep-dives, and case studies coming soon. If you want to see something specific covered, email us.