Do the work from where you already are
Assess a vendor, file evidence, publish a policy, or check a deadline from Claude, Cursor, your terminal, or your own code. Core GRC uses your permissions every time.
- Modules covered
- 5
- With OAuth sign-in
- MCP
- No extra access
- Your role
Ways to connect
Claude
Add the Core GRC connector in Claude on the web, Desktop, or mobile, and work across every module you have.
Learn moreClaude Code and Cursor
Add Core GRC as a remote MCP server in your coding agent.
Learn moreCommand line
Install coregrc from npm to run the same tools in your terminal and scripts.
Learn moreAPI
Call any tool over HTTPS with a read-only or read-and-write API key.
Learn moreCore IQ in the app
Research vendors, run full reviews, and chat over your documents inside the risk suite.
Learn moreEvery tool, listed
See each tool by module, and whether it reads, adds, or changes data.
Learn more
Changes stay safe
Your permissions, nothing more
Connecting doesn't create a new role. A user who can only view vendors gets only the tools that read vendors.
Asks before changing data
Tools that update, publish, archive, or delete are marked so the client asks you first. Read tools run straight away.
A second check on big changes
Deleting, archiving, restoring a version, publishing consent settings, and anything that uses a paid licence also need an explicit confirm.
Every change is labelled
Records changed through MCP are saved with source mcp, and through the CLI with source cli.
Ready to run risk and privacy in one place?
Book a walkthrough of the modules that match your program. Start with one, add more when you need them.