"Basedash AI Kit" isn't an official product name — Basedash doesn't sell a separate kit. What people mean by it is the bundle of AI features built into Basedash, an AI-native business intelligence platform: a chat that answers questions about your data, natural-language dashboards, and a few newer pieces (Automations, Insights, Autopilot, Skills) that only started shipping together in late 2025.
Short answer: There's no standalone "Basedash AI Kit" — it's the set of AI tools inside Basedash's BI platform: an AI data analyst chat, natural-language dashboards, a semantic layer, Automations, Insights, and the Autopilot agent (launched December 2025). Plans start at $1,000/month for up to 25 seats, with a 14-day free trial and no credit card required.
I test AI tools for a living, and this one made me stop and check the naming before writing anything. I tested the claims below against Basedash's own pricing page, its published product docs, and the recorded onboarding walkthrough Basedash puts out publicly, rather than assuming a keyword search implied a real feature. When I worked through that walkthrough step by step, the actual product matched what the docs describe closely enough that I'm comfortable calling this guide accurate — but the "kit" framing needs a correction before anything else, so that's where this starts.
What you'll need
A Basedash account, which starts with a 14-day free trial and doesn't ask for a card up front, according to Basedash’s pricing page. You'll want a real data source to connect eventually — PostgreSQL, MySQL, BigQuery, Snowflake, or one of the 750+ sources Basedash reaches through Fivetran — but you don't need one on day one. Basedash ships a demo database (a "clothing store" dataset) specifically so you can try the AI features before touching production data. Some SQL familiarity helps for judging whether the AI got a query right, though Basedash's own FAQ is explicit that you don't need to write SQL yourself. If you're evaluating this for a small team rather than an enterprise data org, read the pricing section below before you connect anything real — this isn't a $20-a-month tool.
Step-by-step: setting up the Basedash AI kit
1. Start the trial and connect the demo data first
Sign up, then go to Data source → Add data source and pick the built-in demo database rather than your production one. In Basedash's own walkthrough, this is exactly how the onboarding starts — the demo dataset lets you test natural-language charts without any real risk to your data.
2. Build one dashboard with a plain-English chart
Create a dashboard, click New Chart, and type a request instead of writing SQL — "total sales" or "orders by category" is enough. Basedash generates the underlying SQL query and a chart type automatically. In the published walkthrough, that exact prompt produced a KPI card in one step, with the generated query visible if you want to check it.
3. Turn on the semantic layer before more people join
Before inviting a team, define your core metrics once in the semantic layer — what "revenue" or "active user" means — so every chart and every AI answer pulls from the same definition. Skipping this step is the single most common way natural-language BI tools drift: two people ask similar questions and get subtly different numbers because nobody agreed on the metric first.
4. Save repeat questions as Skills
Skills are Basedash's term for reusable AI instructions — save a phrasing or a standard breakdown once, and reuse it instead of re-typing the same prompt every week. This is the closest thing Basedash actually sells to a packaged "kit": a small library of AI instructions your team builds up over time, not something you download.
5. Turn on Automations and Insights for passive reporting
Automations run scheduled AI-powered workflows (a weekly export, a Slack alert on a threshold); Insights sends an automated daily data briefing without anyone opening the app. Both are included on the Startup plan, not a separate add-on.
6. Add Autopilot once the basics are working
Autopilot, which Basedash announced on December 10, 2025, is an AI agent that scans your connected sources on a schedule and proactively surfaces anomalies and trends instead of waiting for someone to ask. I'd treat it as the last thing you turn on, after your semantic layer and permissions are actually right — an agent that proactively reports on badly defined metrics just distributes the wrong number faster.
Example prompts you can copy
These mirror the phrasing Basedash's own product walkthrough uses, which is a good sign the model is tuned for this style of request rather than a generic SQL-generation prompt:
- A quick KPI: "Total sales this quarter" or "How many orders did we get last month?"
- A breakdown: "Orders by product category, last 90 days"
- Refining an existing chart: "Only show products in this category where the price is higher than the selected product"
- A daily briefing (Insights): "Summarize anything unusual in signups, revenue, or churn from the last 24 hours"
- A Skill worth saving: "Whenever I ask about revenue, break it out by plan tier and show week-over-week change"
Common mistakes to avoid
The mistake I'd flag first is treating "AI Kit" as something you can buy separately — it's marketing shorthand for features already inside the base plan, and searching for a standalone SKU wastes time. Second, people underestimate price: Startup plans run $1,000/month for up to 25 seats with $1,000 in AI usage credits included, not a per-user rate like most BI tools, so it's the wrong fit for a two-person shop that just wants a cheap dashboard. Third, skipping the semantic layer before scaling the team causes exactly the metric drift it's built to prevent — set it up before Automations and Autopilot are pulling numbers for other people. Fourth, don't expect direct CSV uploads; Basedash's own FAQ confirms you need a connector like Google Sheets or Drive, since raw file uploads aren't supported. Fifth, deleted charts and dashboards recover from a 30-day trash, but deleted chats don't — there's no undo there.
Tools that make this easier
Basedash is built for teams already running a database, not for someone who just wants an AI chatbot — if that's closer to what you need, my starter kit for AI covers the free chatbot accounts worth opening before spending anything. If you're evaluating BI-adjacent AI tools for a small business budget, my best AI tool for small business roundup covers cheaper, more general options. Developers who'd rather write SQL themselves with AI assistance than hand it to a chat layer should see my best AI tool for code guide instead. Since Basedash's MCP server connects external AI clients to your data, anyone comparing coding assistants for that kind of integration work might also check ChatGPT alternatives for programming. Autopilot's proactive, scheduled-agent design is worth comparing against general-purpose agent tools too — see how to use ChatGPT agents for that angle. And if you want the method behind how I verify claims like these before publishing them, how we test AI tools explains the process.
My take
Basedash is a real, well-documented AI-native BI platform, and the individual pieces — natural-language charts, a semantic layer, Autopilot, an MCP server for outside AI clients — hold up against what the docs and public demo show. What doesn't hold up is the "AI Kit" framing itself; there's no separate product by that name, and anyone searching for one should know that before signing up for anything. At $1,000/month minimum, this is a tool for a team already committed to a shared data stack, not a budget pick, and the free trial is the right way to confirm that before paying.
Frequently Asked Questions
Is the Basedash AI Kit free?
No — there's no separate "AI Kit" to buy, and Basedash itself isn't free past the trial. The Startup plan is $1,000/month for up to 25 seats with $1,000 of AI usage credits included, and a 14-day free trial with no credit card is available to test it first, per Basedash’s pricing page.
Is "AI Kit" an official Basedash feature?
No. It's not a name Basedash uses for any single product. The closest match is Skills — reusable AI instructions you save inside the platform — combined with the AI data analyst chat, Automations, Insights, and Autopilot, which together are what most people mean by the phrase.
How long does it take to set up?
Connecting the built-in demo database and building a first AI-generated chart takes a few minutes, matching Basedash's own published walkthrough. Connecting a real production database, defining your semantic layer, and setting team permissions properly takes longer — plan for a few hours over the first week, not a single sitting.
What's the easiest way to start?
Sign up for the 14-day trial, connect the demo data source instead of a live database, and build one chart with a plain-English prompt before inviting anyone else or connecting real data.
Does it work with AI tools other than Basedash's own chat?
Yes. Basedash ships an MCP server specifically so external AI clients can connect to your data instead of using Basedash's built-in chat exclusively.