FreeYour four-agent team is readyClaim Your AI Team
Docs

Trust and developers

Security

Use deliberate access, appropriate data, and clear human oversight for agent work.

3 min read

Define the boundaries

  • Assign a human owner for important outcomes and sensitive services.
  • Specify restrictions and actions that need approval.
  • Connect only the accounts and sources required for the work.
  • Review boundaries when a job or team responsibility changes.

Share data deliberately

Only provide information you are authorized to use. Separate public, internal, confidential, and regulated data according to your organization’s policies. A provider connection or Computer Use permission may expose more information than one uploaded file; review its scope carefully.

Review consequential actions

Agents can make mistakes or lack essential context. Verify important results and review actions involving money, publication, personal data, access changes, or irreversible effects. A permission request grants only the access described; it is not a substitute for judgment about the job itself.

If you suspect a security issue

  1. 01

    Limit exposure

    Stop the affected job or revoke the relevant access where safe.

  2. 02

    Preserve context

    Record the time, agent, job, and observable issue without copying secrets.

  3. 03

    Report the concern

    Use support@nerova.ai and identify the message as a security report.

  4. 04

    Review affected accounts

    Revoke or rotate provider access when exposure is possible.