Connector performs write actions

Last updated Oct 17, 2026 Preview Policies Access & Sharing

AI summary

Claude Enterprise connectors can perform write actions that create issues, edit documents, or move files in external systems. A Medium severity policy detects which connectors are writing data so you can verify whether the activity is intended.

Severity: Medium Security Recommended Preview

What it detects

Detects connectors that changed data in other systems rather than only reading it, from the Claude Console analytics of the Enterprise plan.

Why it matters

A connector that performs write actions changes data in the other system: it creates issues, edits documents, posts messages, moves files. That is legitimate for an integration the organization chose, and a surprise for one that was expected to only read. The classification comes from the trusted read-only annotation of each tool call, so the absence of write calls is not proof that a connector cannot write - unclassified calls carry no annotation at all. Use this policy to confirm which connectors change data in other systems and whether that is intended.

The rule

Every Claude Connector Usage where Write Calls (30d) is greater than 0

In Rencore Governance

Details

Applies to	Claude connector usage
Connector	Claude Enterprise
Severity	Medium
Category	Security
Recommended	Yes
Recommended automation	None
License type	Standard

Other policies that check Claude connector usage objects:

Connector used with a personal login (Medium)

Frequently asked questions

What does the connector performs write actions policy detect?

The Connector performs write actions policy detects connectors that changed data in other systems rather than only reading it, from the Claude Console analytics of the Enterprise plan.

Which objects does the connector performs write actions policy check?

The Connector performs write actions policy checks Claude connector usage objects from the Claude Enterprise inventory and reports each one that matches its rule as a violation.

How severe is a violation of the connector performs write actions policy?

Violations of the Connector performs write actions policy have Medium severity. The policy is in the Security category.

Which module includes the connector performs write actions policy?

The Connector performs write actions policy comes with the Claude Enterprise connector (in preview) in the AI & Agents module of Rencore Governance, an add-on module on the Premium and Enterprise Plans.

Was this page helpful?