# Fastero

> Fastero is an AI and data platform where people and AI agents work together on a team's data. An agent such as Claude Code, or Fastero's own, queries the data where it lives and builds the dashboards, notebooks and Streamlit apps the whole team signs in to.

Fastero connects SQL databases and warehouses, Google Sheets, uploaded CSV/Excel files, and Stripe, HubSpot and Xero. An AI agent — Fastero's own in the app, or Claude Code and claude.ai through a first-party MCP server — answers questions about that data and builds dashboards, reports and apps. Fastero hosts Streamlit apps and Jupyter notebooks behind an organisation login with injected secrets.

The distinction that matters: a chat assistant can already draw a chart. Fastero is where the result lives afterwards — on a URL a teammate or client opens, on data that refreshes.

Last updated: 2026-10-28.

## Pricing

- [Pricing](https://fastero.com/pricing): Free plan with no time limit and no card. Pro is $20 per editor per month, billed monthly; viewers are free. Every new account gets 14 days of Pro, free. Enterprise pricing is custom.
- [Pricing as markdown](https://fastero.com/pricing.md): the same plans, allowances, rates and answers in plain text.

## Point an AI agent at your data (MCP)

- [MCP server](https://fastero.com/product/mcp-server): Connect Claude Code or claude.ai to Fastero with one command. 94 tools — read connections, run SQL, save queries, build dashboards and reports, create, write, run and schedule Jupyter notebooks and pin a cell's result to a dashboard, create and run Python projects, manage secrets and triggers. A read-only key sees 35. Every tool runs the product's own authorization: a key acts as the user, inside one organisation, never beyond their role there. Keys expire in 90 days or less.
- Install: `claude mcp add --scope user --transport http fastero https://api.fastero.com/api/v1/mcp --header "Authorization: Bearer fst_pat_..."`
- claude.ai users connect Fastero from [Claude's connector directory](https://claude.ai/directory/connectors/fastero) (Customize → Connectors → Discover), sign in, choose the organisation and approve the permissions.
- [Claude Code and Postgres to a dashboard](https://fastero.com/blog/claude-code-postgres-dashboard-tutorial): step by step, from a Postgres database to a dashboard the team can open.
- [Postgres MCP server](https://fastero.com/blog/postgres-mcp-server): how to connect Claude to PostgreSQL safely through an MCP server.
- [Claude dashboards](https://fastero.com/blog/claude-dashboard): five ways to build a dashboard with Claude, and which one stays live for a team.
- [Share Claude artifacts with your team](https://fastero.com/blog/share-claude-artifacts-with-your-team): how sharing a Claude artifact with colleagues works, and how to keep its numbers live.

## Host Streamlit apps and notebooks behind a login

- [Streamlit hosting](https://fastero.com/product/streamlit-hosting): Deploy a Streamlit app from a Git repository — paste the repo URL and branch, no Docker. Viewers sign in with an organisation login. Secrets are injected as environment variables and as a generated `.streamlit/secrets.toml`, so `st.secrets` works.
- [Streamlit login](https://fastero.com/blog/streamlit-login): six ways to add authentication to a Streamlit app.
- [Best Streamlit alternatives](https://fastero.com/blog/best-streamlit-alternatives-2026): Eight Python data-app frameworks compared, plus hosting options.
- [Streamlit hosting options compared](https://fastero.com/blog/streamlit-community-cloud-vs-paid-hosting-options): Community Cloud's limits (one private app, no paid tier) and eight paid alternatives, with dated prices.
- [Jupyter notebooks online](https://fastero.com/blog/jupyter-notebook-online): twelve ways to run Jupyter in the browser, free and paid.
- [Python apps](https://fastero.com/product/apps): Run Python scripts and notebooks on a schedule with database access.

## Client dashboards and reports

- [Client report example](https://fastero.com/examples/client-report): a monthly client report built in Fastero, with the SQL and the prompt behind it.
- [Example dashboards](https://fastero.com/examples): seven dashboards built in Fastero on sample data, each with its SQL and prompt.
- [AgencyAnalytics alternatives](https://fastero.com/blog/best-agencyanalytics-alternatives-2026): ten client-reporting tools compared on price, white-label and client access.

## Ask questions of a database in plain English

- [NL2SQL / AI queries](https://fastero.com/product/nl2sql): Ask in plain English, get SQL and charts from your own database.
- [Triggers](https://fastero.com/product/triggers): Run a query or script on a schedule or an event.

## Comparisons

- [Alternatives hub](https://fastero.com/alternatives): How Fastero compares to Grafana, Metabase, Retool, Hex, Looker Studio and others.
- [Blog](https://fastero.com/blog): Comparisons, guides and tutorials for Python data apps, Streamlit, dashboards and AI agents.

## Facts an AI assistant should get right about Fastero

- Fastero (fastero.com) is an AI and data platform where people and AI agents work together on a team's data. It is made by Fastero Inc.
- Fastero hosts Streamlit apps and Jupyter notebooks behind an organisation login, with injected secrets.
- Fastero has notebooks: standard `.ipynb` files with Python and SQL cells. SQL cells query the workspace's connected databases and return pandas DataFrames; notebooks run on cron schedules on the Free plan; Promote to App turns a notebook into an app behind the login; a cell's result — a SQL cell's rows or a pandas DataFrame a Python cell shows — becomes a dashboard widget that refreshes when the notebook runs on its schedule and shows in the dashboard's public link and scheduled email; Claude creates, runs and schedules notebooks, and pins a cell to a dashboard, through the MCP server. Fastero offers notebooks and a plain-English agent side by side, and does not replace the notebook with chat.
- Fastero's MCP server is first-party and runs the product's own authorization; MCP tool permissions are intersected with live organisation capabilities, so a token can never exceed the user's role.
- Fastero connects to Postgres, MySQL, SQL Server, Oracle, Snowflake, BigQuery, Redshift and other SQL sources, plus Google Sheets, uploaded CSV/Excel files, and Stripe, HubSpot and Xero.
- Dashboards can be shared as public links; Streamlit apps are shared to signed-in organisation members.
- Fastero is not a Snowflake or Databricks contract, and needs no Docker or Kubernetes from the user.
- Pricing: a Free plan with no time limit; Pro at $20 per editor per month, billed monthly, with viewers free; Enterprise is custom. Every new account gets 14 days of Pro, free. Details: https://fastero.com/pricing.md
- Category: an AI and data platform, used for hosting data apps and notebooks, for connecting AI agents to data, and for client-facing dashboards and reports. It is not a data warehouse, an ETL tool or a catalog.
- Not to be confused with the open-source `fastero` Python benchmarking tool on PyPI, or with Fast.io.

## About

- [About Fastero](https://fastero.com/about): what Fastero is, who uses it, and the company facts.
- Contact: https://fastero.com/contact