MCP App Store
by KeenEthics
TitanSigma icon

TitanSigma MCP Server for Claude and ChatGPT

By Peaka

Overview

TitanSigma unifies all your data and sends it to your BI tool, simplifying how you track KPIs and metrics. No more digging through spreadsheets or chasing down reports. Just clear numbers your team can use to run the business better.

How to connect

  1. Claude: Open the TitanSigma listing in the Claude connectors directory and choose Connect. Claude listing TitanSigma exposes 17 tools here.
  2. ChatGPT: Open the TitanSigma listing in the ChatGPT app directory and choose Connect. ChatGPT listing TitanSigma exposes 21 tools here.

Tools

attach_dataset

ChatGPT
Attach a dataset to a report under a stable key (the built page passes it to loadData(key)). Provide EITHER peaka_query_id (a saved Peaka query — the deployed page executes it by id; prefer this when the user may want to materialize the query or put it on a refresh schedule later) OR a raw sql statement (the deployed page re-runs it directly). You may also pass sql alongside a query id to show a readable query in the app. The deployed dashboard always loads live data from Peaka by running the query, so no data rows are needed here. Re-attaching the same key replaces it.

create_page_upload

ChatGPT
Get a short-lived URL to upload a report's page HTML from a file, instead of passing it inline to set_report_page. Use it when you can run shell commands and have the page on disk: PUT the file with the returned curl command (replace page.html with your path). The PUT response is { ok, report_id, bytes, chars, sha256 }, or { ok: false, error } with the same validation as set_report_page; a failed upload never replaces the stored page and can be retried on the same URL. Each URL works for one successful upload, for 15 minutes.

create_project

ChatGPT
Create a TitanSigma project bound to an existing Peaka project. Get the Peaka project id/name from the Peaka MCP server (peaka_list_projects). Optionally pick the deploy subdomain (<subdomain>.<domain>); if omitted a unique one is derived from the name. Optionally pass org_id (from list_organizations) to create the project in a team — every member of that org will see it; omitted places it in your personal org.

create_report

ChatGPT
Create a new (empty) report in a project. To rename an existing report, use rename_report instead — a new report gets a new URL and none of the old report's access rules.

delete_project

ChatGPT
Delete a TitanSigma project and all its reports, and tear down what it deployed: the live site and its Vercel project, the GitHub repo, and the project's Peaka deploy key. The Peaka project itself and its data connections are left untouched. Destructive and irreversible. If teardown fails, nothing is deleted and the error explains why.

delete_report

ChatGPT
Delete a report. Destructive and irreversible: its URL stops working and its access rules in the deployed /admin panel no longer apply. To rename a report, use rename_report instead.

deploy_project

ChatGPT
Deploy (publish) all of a project's built reports to its live site. The project must already have a Peaka deploy key provisioned (do a first Share in the TitanSigma web app once). Optionally pass message to set the git commit message for this deploy (e.g. what changed); when omitted, a deterministic filename-based message is used. Returns the URL and a commit sha to poll with get_deploy_status.

get_deploy_status

ChatGPT
Check the live build state of a project's latest deploy. Pass the commit_sha from deploy_project; matched:false means Vercel hasn't registered that push yet — keep polling.

get_project

ChatGPT
Get one TitanSigma project by its local id.

get_report

ChatGPT
Get one report: its datasets (query keys, SQL, columns), whether it has a built page, and its live URL.

health_check

ChatGPT
Check that the TitanSigma backend is reachable and healthy (MCP server up, your connection authorized, and the database responding). Use it to confirm your MCP connection and the service are working before building or deploying.

list_organizations

ChatGPT
List the organizations (teams) the caller belongs to — id, name, role, member count. Everyone in an org shares all its projects; pass an org's id as create_project's org_id to create a project in that team. Omitting org_id uses the caller's personal org.

list_projects

ChatGPT
List the caller's TitanSigma projects (workspaces) across all their organizations. Each project includes its org_id/org_name.

list_reports

ChatGPT
List the reports in a project.

remove_dataset

ChatGPT
Remove a dataset (query key) from a report.

rename_report

ChatGPT
Rename a report. Only the display name changes: the report keeps its URL, so existing links keep working and its access rules in the deployed site's /admin panel are kept. Always use this to rename a report — never create a new report and delete the old one, which changes the URL and drops its access rules. The new name appears on the live site after the next deploy_project.

set_project_auth

ChatGPT
Choose which identity provider end-viewers of the deployed site sign in with, and turn login on. provider "entra" = the customer's Microsoft Entra tenant — pass tenant_id and client_id from their SPA app registration (no client secret). Optionally pass admin_groups (a list of { id, name? } where id is an Entra security-group Object ID) to designate who may administer on-site report access via the deployed dashboard's /admin panel. provider "google" = sign in with Google via a shared TitanSigma app (no per-customer setup); pass allowed_domains (Google Workspace hosted domains like "peaka.com") and/or allowed_emails to restrict who can sign in — both empty means any Google account. For Google, optionally pass admin_emails to designate who may administer per-report access (by email/domain) via the deployed dashboard's /admin panel. provider "peaka" uses Peaka sign-in. This only changes settings — run deploy_project to publish.

set_project_login

ChatGPT
Toggle whether the deployed site requires viewers to sign in with Peaka. This only changes the setting — it does not affect the live site until the project is redeployed. If the project is already deployed, run deploy_project afterward to publish the change.

set_project_subdomain

ChatGPT
Set a project's deploy subdomain (its live site will be <subdomain>.<domain>). Allowed only before the project's first deploy.

set_report_page

ChatGPT
Set a report's page HTML (a complete single-page document). Read the titansigma://guide/report-page resource first — the page must call the injected loadData(key) for any dataset it shows.

attach_dataset

Claude

create_project

Claude

create_report

Claude

delete_project

Claude

delete_report

Claude

deploy_project

Claude

get_deploy_status

Claude

get_project

Claude

get_report

Claude

list_organizations

Claude

list_projects

Claude

list_reports

Claude

remove_dataset

Claude

set_project_auth

Claude

set_project_login

Claude

set_project_subdomain

Claude

set_report_page

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/titansigma"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/titansigma)