Corebasehq
corebasehq.com
As featured in
Follow
Overview
News
Technologies
Salaries
Products
People
Growth
Financials
Overview
Give your AI agent real access to real systems without giving it unlimited access — scoped per caller, human approval where it matters, every call on the record. Query your databases, REST/GraphQL APIs, and 50+ apps with an OpenAI-compatible API, or embed the chat widget.
News
GitHub - CoreBaseHQ/coremcp: CoreMCP: Connect Legacy Databases to AI Agents via Model Context Protocol. Open-source bridge for LLM data analysis.
CoreMCP: Connect Legacy Databases to AI Agents via Model Context Protocol. Open-source bridge for LLM data analysis. - CoreBaseHQ/coremcp
Read more
Report
Your AI Agent's Chat History Is User Input
Everyone sanitises the user's message. Almost nobody sanitises the assistant's - and that's the one the model believes.
Read more
Report
Prompt Injection Is an Authorization Problem
You can't filter your way out of this. The tool your agent can't see is the only one no message will ever talk it into calling.
Read more
Report
Human Approval Is a Privilege Escalation Path
Everyone builds agent approvals as a boolean that skips the check. That turns a person's click into the authorization itself, find something the agent was never allowed to do, get a human to say yes, and the yes is now the permission.
Read more
Report