CoreBase
Regulated infrastructure for AI agentsBuild agents with audit and controls
Highlights
What you can do with it
- Let staff query a SQL Server database in natural language.
- Embed a governed assistant inside project management software.
- Run scheduled multi step data collection across SaaS tools.
How to use it
- 1 Register an account and start on the free plan.
- 2 Connect databases, APIs and the SaaS apps you use.
- 3 Set permissions, read and write rules and cost controls.
- 4 Build agents with the compatible API or the chat widget.
- 5 Test, deploy, then monitor through the audit logs.
Who it is for
Pros
- Free plan to start without a credit card.
- Audit and permission model suits regulated work.
- Compatible API means less rework.
- Bridge reaches legacy and on premise systems.
Limitations
- Enterprise setup is a real project, not a weekend.
- Governance rules need ongoing maintenance.
- Self hosted bridge adds operational work.
- Free plan limits and pricing can change.
What CoreBase is
CoreBase is regulated infrastructure for building and delivering AI agents that customers can trust. The pitch is that most agent tools are easy to demo and hard to govern, so CoreBase puts the controls in first. It ships built in connectors, permission management, audit logs and cost control, plus an OpenAI compatible chat API and an embeddable chat widget you can drop into your own product. Agents can query and act on databases, REST and GraphQL APIs and more than 50 applications, with multi tenant isolation through row level security and full audit logs. Database sessions are read only by default, with source based write governance on top. An open source bridge called CoreMCP lets agents reach local and legacy systems such as SQL Server 2000 and ERP. Free, professional and team plans are offered, with enterprise pricing customised.
What you can do with it
- Query databases and SaaS apps in natural language.
- Embed a governed assistant in your own product.
- Run multi step automations with approval gates.
- Keep every action inside an auditable permission model.
Who it is for
- Enterprise developers shipping agent powered products.
- Managers who need cost control and tenant isolation.
- Data analysts who want natural language access to data.
What to watch out for
This is enterprise infrastructure, so a serious deployment is a project involving security review, data modelling and rule design rather than an afternoon. Governance rules need ongoing maintenance as systems and staff change. Running the bridge for legacy access adds operational work on your side. The free plan is a starting point, and free limits, paid tiers and enterprise pricing can change.
FAQ
What is CoreBase for?
Building and delivering AI agents with the connectors, permissions, audit and cost control that enterprises expect.
Does it work with existing SDKs?
Yes, it offers an OpenAI compatible chat API so existing SDKs can point at CoreBase.
Is there a free plan?
Yes, free, professional and team plans are offered, and the free plan starts without a credit card.
Can it reach old systems?
Yes, the open source CoreMCP bridge lets agents access local and legacy systems such as SQL Server 2000 and ERP.
Why it is free
What the official page says： 免费
Free limits, trial length and features are decided by the official site and can change at any time.
Product preview
Screenshot archived from the listing page, for preview only.
