MCP Server — Agent Integration

Tabular Pro exposes a hosted Model Context Protocol (MCP) endpoint that AI agents (Claude, GPT, Cursor, and others) can plug into directly. This enables powerful automated workflows like:

An agent that monitors survey responses and alerts you to trends
A research assistant that analyzes open-ended feedback on demand
Automated report generation from survey data
CI/CD pipelines that create surveys programmatically

What is MCP?

The Model Context Protocol is an open standard for connecting AI agents to external tools and data sources. MCP servers expose "tools" that agents can discover and call.

Endpoint

https://tabularpro.com/api/mcp

Nothing to install. The server is hosted and always up-to-date — point any MCP-compatible client at the URL with a Tabular Pro API key.

Authentication

Generate an API key from Settings > API Keys in the Tabular Pro app, then send it as a bearer token on every request:

Authorization: Bearer tp_your_key_here

The X-API-Key: tp_your_key_here header is also accepted.

Using with Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "tabularpro": {
      "url": "https://tabularpro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer tp_your_key_here"
      }
    }
  }
}

Using with Claude Code

Add to your project's .claude/settings.json:

{
  "mcpServers": {
    "tabularpro": {
      "url": "https://tabularpro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer tp_your_key_here"
      }
    }
  }
}

Using with Cursor / other clients

Any client that supports remote MCP (Streamable HTTP transport) works the same way — provide the URL and an Authorization header.

Available Tools

Surveys & Questions

list_surveys, get_survey, create_survey, update_survey, delete_survey, duplicate_survey, publish_survey, list_questions, create_question, update_question, delete_question, list_pages, create_page

Responses & Data

get_response_stats, get_response_grid, get_question_data, export_responses, get_distribution

AI Analysis

run_sentiment_analysis, run_theme_extraction, run_text_classification, generate_insights, summarize_responses, detect_anomalies, translate_responses

Analysis

get_frequency_analysis, get_summary_statistics

Dashboards & Reports

list_dashboards, get_dashboard, create_dashboard, get_chart_data, list_reports, get_report, export_report_pdf

Dashboards contain pages with elements. Element types include charts (bar, horizontal-bar, stacked-bar, line, area, pie, donut, radar, gauge, kpi, line-bar, table) and non-chart elements: text (annotations), filter (interactive dashboard filter control), and variable (inline computed value). See the Dashboards API reference for config shapes.

Organization

get_current_user, list_users, list_workspaces, get_workspace, get_overview_stats, search, list_templates

Field Operations

list_waves, get_wave_progress, list_survey_links

Example Workflow

You: "Summarize the feedback from our Q2 satisfaction survey"

Agent:
1. list_surveys → finds "Q2 Customer Satisfaction" (id: 42)
2. get_response_stats(42) → 1,247 complete responses
3. list_questions(42) → finds open-ended Q5 "Any additional feedback?"
4. summarize_responses(42, question_id: 5) → AI-generated summary
5. run_sentiment_analysis(42, question_id: 5) → 72% positive, 18% neutral, 10% negative
6. generate_insights(42) → key findings and actionable recommendations

The agent handles all the API calls automatically — you just ask natural language questions about your data.