Syntitan

Syntitan MCP Server The official Model Context Protocol (MCP) server for Syntitan by CUBIG — the AI-Ready Data Platform that fills the missing layer between…

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

  • List Datasets: List datasets the user can access, with each one's latest version, row/column counts, AI Readiness score, and last-modified time. Timestamps are ISO 8601 in UTC; when presenting, infer
  • Get Dataset History: Return a dataset's full version history, including each version's release status, author, timestamp, and changes (masked PII entities and applied refinement modules). Timestamps a
  • Get Ai Readiness Summary: Return a dataset's overall AI Readiness score with per-category scores (Integrity, Contextuality, Conciseness) and the top-priority issues. For metric definitions and interpr

What data it sees

Do you need an account

No: the server works without sign-in

Syntitan MCP Server

The official Model Context Protocol (MCP) server for Syntitan by CUBIG — the AI-Ready Data Platform that fills the missing layer between enterprise data management and AI execution. It lets Claude and other MCP-compatible clients (Cursor, VS Code, Claude Code, ChatGPT, etc.) securely connect to Syntitan through natural language.

  • Category: Data & Analytics
  • Transport: Streamable HTTP
  • Authentication: OAuth 2.0
  • Endpoint: https://mcp.syntitan.ai/mcp

Connecting

Add Syntitan as a remote MCP server in your client using the endpoint URL below:

https://mcp.syntitan.ai/mcp

On first connection your browser opens the Syntitan sign-in page. Log in and authorize access; the client then connects automatically. No API keys are stored in the client.

Cursor, VS Code, Claude Code, and other MCP clients support remote MCP servers — add the endpoint URL above following each client's connector instructions.

Tools

  • List Datasets — List datasets you can access, with version, row/column counts, AI Readiness score, and last-modified time
  • Get AI Readiness Summary — Overall AI Readiness score with per-category scores (Privacy, Conciseness, …) and top-priority issues
  • Get AI Readiness Detail — Break down a specific category to show which columns contributed and by how much
  • Get Dataset Schema — Column structure (type, missing values, PII flags) without exposing PII values
  • Get Column Profile — Detailed statistics for a single column (distinct/top values, missing rate, examples)
  • Get Dataset Sample Rows — Sample rows with PII auto-masked; refuses if unmasked PII is present
  • Get Dataset History — Version history (release status, author, timestamp, changes)
  • Get Diagnosis Guide — Guidance on interpreting AI Readiness diagnostic metrics
  • Get Refinement Options — Available preprocessing options (modules, de-identification targets)
  • Estimate Refinement Credits — Preview estimated credits and time before running a refinement
  • Start Refinement — Run preprocessing/de-identification (consumes credits); saved as a new snapshot
  • Get Refinement Status — Progress/status of a running or completed refinement

Example use cases

  1. "What datasets do I have right now?" — Lists your datasets with their latest version, row/column counts, AI Readiness score, and last-modified time.
  2. "Give me a full diagnosis of this dataset." — Returns the overall AI Readiness score with a per-category breakdown (Privacy, Conciseness, Contextuality, …) and the highest-priority issues to fix first.
  3. "Which columns have the missing values, and why?" — Drills into a specific category to show exactly which columns contributed and by how much.
  4. "Show me 10 sample rows where region is null." — Returns sampled rows; PII columns are returned only in masked form, and the request is refused if the dataset still has unmasked PII.
  5. "Mask all the PII columns in this dataset." — Runs Syntitan's de-identification preprocessing in place and reports the before/after AI Readiness change.

Authentication & security

Authentication uses OAuth 2.0. Users sign in through the Syntitan website; credentials are never shared with the MCP client. All traffic is served over HTTPS/TLS.

Data handling

Syntitan accesses only the data required to fulfill a user's request. Audit logs (which may include user identifiers such as email) are retained for operational and security purposes in accordance with our privacy policy. Data collection, storage, retention, and deletion practices are described in the privacy policy linked below.

Privacy & terms

Support

License

All rights reserved.

How to connect

How to connect Syntitan to Claude

The server address to paste into Claude settings:

https://syntitan-mcp--cubig.run.tools

  1. Open Claude (claude.ai or the desktop app).
  2. Go to Settings → Connectors.
  3. Click Add custom connector.
  4. Paste the server address copied below into Remote MCP server URL and click Add.
  5. In a chat, click + → Connectors and switch the new connector on.

Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

Authorization

After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

How to connect Syntitan to ChatGPT

The server address to paste into ChatGPT settings:

https://syntitan-mcp--cubig.run.tools

  1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
  2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
  3. Open Settings → Connectors and click Create.
  4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
  5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
  6. In a new chat open + → Apps/Connectors and enable the connector.

OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

Authorization

On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

How to connect Syntitan to Cursor

The server address to paste into Cursor settings:

https://syntitan-mcp--cubig.run.tools

Fastest: click Open in Cursor below and confirm the prompt.

Manually:

  1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
  2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
  3. Save the file. The server appears in the MCP list; authorize it there if asked.

Authorization

If the service needs sign-in, an authorize button appears next to the server in the MCP list.

Server tool list (12)

Raw names from tools/list. Only developers need these.

list_datasetsList datasets the user can access, with each one's latest version, row/column counts, AI Readiness score, and last-modified time. Timestamps are ISO 8601 in UTC; when presenting, infer the user's timezone from conversation context and convert, otherwise show UTC. Results are paginated (`total` = full count).
get_dataset_historyReturn a dataset's full version history, including each version's release status, author, timestamp, and changes (masked PII entities and applied refinement modules). Timestamps are ISO 8601 in UTC; infer the user's timezone from context and convert, otherwise use UTC.
get_ai_readiness_summaryReturn a dataset's overall AI Readiness score with per-category scores (Integrity, Contextuality, Conciseness) and the top-priority issues. For metric definitions and interpretation, call `get_diagnosis_guide`.
get_ai_readiness_detailBreak down a specific AI Readiness category to show which columns contributed and by how much. category: 'integrity' (nulls/outliers/type-mismatches), 'contextuality' (semantic coverage), 'conciseness' (unique values/redundancy). For metric definitions, call `get_diagnosis_guide`.
get_dataset_schemaReturn a dataset's column structure — type and missing values — without exposing PII column values. For metric definitions and interpretation, call `get_diagnosis_guide`.
get_column_profileReturn detailed statistics for a single column, such as distinct/top values, missing rate, and example values. Returns statistics only (no raw values). For metric definitions and interpretation, call `get_diagnosis_guide`.
get_dataset_sample_rowsReturn a sample of rows from a dataset with PII columns auto-masked; refuses if unmasked PII is present. filters: list of {column, op, value}; op one of eq|ne|gt|gte|lt|lte|is_null|not_null|contains (AND). Returns up to 50 rows plus total_matched. Unmasked PII columns are excluded (listed in `excluded_columns`); filtering on a PII column is rejected.
get_diagnosis_guideReturn guidance on how to interpret AI Readiness diagnostic metrics and scores. Call this before interpreting diagnosis results to avoid common misreadings. No arguments.
get_dataset_run_optionsReturn the available preprocessing options for a dataset, such as preprocessing modules and de-identification targets. Includes recommended/disabled flags and detected PII entity types eligible for redaction (redaction: {entities: [detected uppercase entity keys], available: bool}). Call this first before estimate_dataset_run or start_dataset_run to discover valid parameter values.
estimate_dataset_runPreview the estimated credits and time required before running a refinement. Call get_dataset_run_options first to see valid refinement_modules and redaction_entities values, then start_dataset_run to execute.
start_dataset_runRun preprocessing/de-identification on a dataset (consumes credits); results are saved as a new snapshot. Call get_dataset_run_options first to discover valid module and entity values, then estimate_dataset_run to preview cost. Because this charges credits, ALWAYS show the user the estimated cost and the modules/entities to be applied, then get their explicit confirmation before calling (a natural confirmation in the conversation, not a rigid Y/n prompt). Returns a run_id; poll get_dataset_run_status to track progress.
get_dataset_run_statusReturn the progress and status of a running or completed refinement. Use the run_id returned by start_dataset_run; poll until status is 'completed' or 'failed'. On completion, returns before/after AI Readiness scores and trust breakdown (integrity, contextuality, conciseness).