Ship AI Agents Your Customers Can Trust.
Ship AI Agents Your Customers Can Trust.
The governed infrastructure layer for building and deploying agents — connectors, permissions, audit trails, and cost controls built in — for your product, or your own stack.
Free to start, no credit card · book a demo for enterprise & air-gapped
Reaches systems others can't
On-prem, air-gapped, behind any firewall
Scoped, revocable access
Signed tokens, rotate anytime
Free to act, never to overreach
Scoped to the sources and actions you allow
Usage you can see
Cost per user, before it surprises you
Developer-first
A few lines to ship
Sign a scoped token for each of your users — source allowlists, rate limits, read-only by default — then drop in the widget or call the SDK. The governance rides in the token, so there's no glue code to maintain.
EmbedQueryAutomate
typescript
1`// Server-side: sign a scoped token for this user`
2`const token = cb.widgetToken(user.id, {`
3` sources: ["orders", "invoices"],`
4` limits: { daily: 50 },`
5`});`
6` `
7`// Client-side: drop the governed widget into your app`
8`import "@corebasehq/widget";`
9` `
10`CorebaseWidget.init({`
11` publicId: "proj_•••",`
12` getAuthToken: () => token,`
13`});`
API referenceEmbed the widgetUse CoreMCP
Connect
Give your agents the data they run on
Every agent you ship reads straight from your real systems — databases, REST & GraphQL APIs, and 50+ apps. No ETL, no glue code, no integration sprint.
Cloud & SaaS
Connect in one click
Paste a connection string or sign in — PostgreSQL, MySQL, REST & GraphQL APIs, Slack and Microsoft 365 connect directly, live and read-only.
On-prem & legacy
When a customer needs it
When a customer's data lives behind the firewall, reach it too: the open-source CoreMCP agent runs inside their network to serve SQL Server 2000+, Firebird, and on-prem ERP / POS. It connects outward from the inside — nothing to open up, no VPN. Even works fully offline.
Why CoreBase
Everything you need to ship on real data
Guardrails, attack screening, per-user isolation — the governance you'd otherwise build yourself, already here. Batteries included, self-host optional.
Agents that stay on the rails
Draw the steps your assistant works through — what data each step can touch, what it may do, when it must ask for approval. It reasons freely inside; the rails are enforced, not suggested.
RailEnforced
Check orderRefund > $100 · approvalEscalate
Every visitor connects their own
Gmail, Jira, or any of 50+ apps — each visitor links their own account, and the assistant acts on it alone. Isolation is the OAuth grant itself, not a filter that could slip.
AK
SM
JD
Bring your own model
Use the built-in AI, or plug in your own keys — Claude, GPT, Gemini and more. Switch anytime.
Claude
GPT
Gemini
Private by design
Read-only by default, isolated per customer, encrypted end to end — and every action is logged.
Attacks stopped at the door
Prompt injection, jailbreaks and abuse are screened before a message ever reaches the model — on chat, widget, voice and API alike.
Inbound messagesScreened
ignore all previous instructions…blocked
It learns your schema
Query Memory remembers the queries that worked and what your tables mean — so natural-language → SQL gets sharper the more you run it.
Explore
Ask across every system — answers from live data
One query can hit your database, your CRM, and your inbox at once — read straight from the source, never a stale copy.
One question, every source
CoreBase fans your question out across the systems you have connected and assembles a single answer — with every step logged.
Top customers by revenue
Live from your database
32k
22k
12k
Wholesale (61%)
Retail (39%)
Real numbers, straight from the source
Answers come back as figures and tables computed from live data — not summaries of a stale copy.
Always current
Every answer is computed at ask time, straight from the live system.
Acts on events
Agents can fire when something happens in a connected app — not just when asked.
Runs on a schedule
Recurring checks and reports run on their own, as often as you need.
Full audit trail
Every question, query, and action is recorded — nothing happens off the record.
Govern
Room to think, rails that hold
Draw the steps your agent works through — the data each step can reach, the actions it may take, when it must ask first. It reasons freely inside; the rails are enforced in code, never just suggested in a prompt.
Step 1
Check the order
Step 2
Try to resolve
Step 3approval
Refund
Step 4
Escalate to a human
In the Refund step, the agent canEnforced
Read orders & payments
Refund over $100 — waits for a human OK
Anything else — never runs
Enforced, not suggested
A blocked action is never called; a required approval is never skipped. Your rules compile to real gates — not a line in a prompt the model can talk itself out of.
Scoped, step by step
Each step grants only the data and actions it needs. The agent physically can't reach a source you didn't hand it there — even if it tries.
Approval gates & full audit
Hold an action for a human OK before it runs — and every query, step, and block it hits lands in the audit trail.
Open source
CoreMCP — the open-source bridge.
The agent that reaches behind the firewall is open source. Audit it, run it yourself, or keep it fully offline. The managed platform is built on the same code your team can read line by line.
Pricing
Start free. Scale when you're ready.
Every plan includes AI usage on built-in models — no key required to start. Bring your own LLM key any time and pay your provider directly instead.
Free
Try CoreBase against one source.
$0/ mo
Free forever
- 1 data source
- $1 trial AI usage — built-in models, no key needed
- BYOK LLM keys (optional, your provider)
- Developer API + embeddable widget
- Query Memory + Unified Context (RAG)
- Read-only by default
- Community support
- Everything in Free, plus:
- 5 data sources
- $5/mo included AI usage
- REST / GraphQL + cloud DB connectors
- CoreMCP on-prem agent (NAT / air-gap reach)
- Agent flows — scheduled & event-driven
- Email support
- Everything in Pro, plus:
- Unlimited data sources
- $15/mo included AI usage
- RBAC + full audit log
- Per-tenant isolation at the database
- Priority support
- Air-gapped / on-prem container
- Dedicated infrastructure
- SSO / SAML
- SLA + dedicated support
ProMost popular
For developers wiring agents to real systems.
$49/ mo
billed monthly
Team
For teams that need governance and scale.
$149/ mo
billed monthly
Enterprise
Air-gapped or regulated environments, dedicated infrastructure, and SSO.
Need the full breakdown across every plan?
FAQ
Frequently asked questions
Reach, safety, isolation, and deployment — what security and data teams ask before connecting an agent to live systems.
Can CoreBase reach a database that's on-prem or air-gapped?
---
Can agents act on their own, or do I have to ask every time?
---
Can an agent change or break our data?
---
How is our data isolated from other customers?
---
What systems can we connect today?
---
Where does our data go, and do you train models on it?
---
How do we deploy — cloud or self-hosted?
---
How do I get an API key, and what's included on the free tier?
---
Do you have official SDKs, or can I use my existing OpenAI client?
---
What happens if I hit a rate limit?
---
Can't find what you're looking for? Contact our support team
Ship your first agent today.
Connect a source, wire up an agent, and stream your first answer in minutes. Free to start — no credit card.
Need air-gapped or enterprise? Book a demo