Basedash is the AI-native Business Intelligence platform that lets you create dashboards and understand your customers using natural language. Connect your data, describe the chart you want, and let AI generate the visualization. No SQL required.
This is the 17th launch from Basedash. View more
Basedash Skills
Launching today
Skills are reusable bundles of instructions that every Basedash AI surface can read on demand.
Define your metrics once and any AI agent in your workspace will pick up the skill when it's relevant. No more pasting the same caveats into every prompt.
Each skill is a short, plain-language playbook for one concept. Admins manage them; everyone else's AI gets the benefit. Add them as you go — the more you teach Basedash, the more it acts like an analyst who already knows your business.
Free Options
Launch Team
Basedash
@maxmusing Hey Max, congrats on shipping Skills 🎉
The "lightweight semantic layer rather than a system prompt" framing is the right one. Most teams reinvent this badly as a growing pile of prompt caveats that nobody owns. Pulling definitions into durable, admin-managed context is the actual fix.
Question on the governance side: what happens when skills drift or conflict? Say an admin defined "activation rate" six months ago, the business changed how it counts it, but the old skill is still live. Or two skills define "active user" slightly differently. Does Basedash surface the conflict, version skills with a clear "current" pointer, or does the most recently fetched one just win? Asking because the value of a semantic layer is only as good as its freshness, and that is exactly where these systems quietly rot.
The visible tool call in the thinking trace is a great touch by the way. "Reading Activation rate skill" before answering is the difference between trust and black box.
Basedash
Thanks @artem_fedorovich! It's up to the user to manage their skills, but the AI will intelligently work through conflicts and ask the user for clarification if it needs additional context. The AI can also pull context from your existing dashboards to understand how your team is actually calculating metrics that are being referenced.
Moving definitions out of one-off prompts and into shared durable context is exactly the lightweight semantic layer most teams skip until they're already drowning in inconsistent metric defs. How do you handle drift when someone updates a Skill that's been silently feeding 12 different surfaces — version pin, broadcast, or both?
Basedash
@eran_shayshon totally agree, and soon we'll allow the AI to manage its own skills automatically so you won't even have to think about it. If you anyone updates a skill we make it easy to broadcast changes with our AI agent. You just tell it "I updated our NRR skill, please update all charts referencing this metric to use the new formula" and it will update everything for you.
this is a very handy tool for founders, but I have a question on how are you handling large dataset like clickstream locally and does the data site locally even after ETL?
Basedash
Hey @harshalvc_ai we can either connect directly to certain sources like Postgres, BigQuery, PostHog, or we can ETL data from external systems into a hosted Basedash Warehouse that we manage for you. Either way we can handle large datasets securely and performantly so that our AI agent can work efficiently.