The 5 Mistral AI security checks Black Cat runs
Black Cat SSPM evaluates 5 security policies against your Mistral AI configuration on every scan, classifies each finding by risk, and provides remediation steps. Browse them by topic below.
How to connect Mistral AI — what access Black Cat needs, and why.
Configuration hardening (3)
- Agent Uses Code Interpreter severity: medium
- Agent Deployed Without Guardrails severity: low
Other checks (2)
severity: high Library Shared Org-Wide With Documents fix difficulty: medium #
Restrict org-wide access to libraries that hold documents
- Open the library in the Mistral console and review its sharing
- Replace the org-wide share with specific users or workspaces
Satisfies: NIS2 Directive NIS2-21.a.3 DORA (SaaS Security) DORA-9.12
severity: medium Library Shared Org-Wide Editable fix difficulty: easy #
Downgrade org-wide Editor access to Viewer or scope it
- Review who needs write access to the library
- Change the org-wide share role from Editor to Viewer, or scope it
Satisfies: NIS2 Directive NIS2-21.a.3 DORA (SaaS Security) DORA-9.12