From the Palo Alto Networks edition of November 2, 2026
Anthropic admin guide: approved EDR sensors, including Palo Alto Networks, can run inside Claude Cowork's local VM in beta; cloud sessions are not covered and sensors can see prompts and files
Anthropic (Claude Help Center) · AI platform administration and endpoint security · Global product documentation
- Publisher
- Claude Help Center
- Original publication
- Claude Help Center article, page shows "Updated today" when retrieved Nov. 2, 2026
- Source retrieved
- 2026-11-03
What happened
Anthropic's documentation, independent of Palo Alto, says running an EDR sensor inside the Cowork VM is in beta for Team and Enterprise plans on macOS and Windows, including Claude Desktop connected to Amazon Bedrock, Google Cloud Vertex AI or Microsoft Foundry. Claude Desktop copies the vendor package into the VM at each start, checks its signature against an approved vendor key, and runs the install script. Palo Alto Networks is one of seven approved vendors. Only the VM on the user's computer is covered; cloud-run Cowork sessions cannot host a sensor. The sensor can see prompts, files and Claude responses and sends them to the EDR vendor, outside the Cowork network allowlist. Anthropic charges nothing extra during beta; package signatures expire within two years.
Why it matters
This is the platform operator's account of the constraints behind Palo Alto's announcement and is the more useful planning document for public-sector admins. The cloud-session gap and the fact that prompts and files flow to the EDR vendor matter for agencies subject to records, FERPA, HIPAA or CJIS rules. Interpretation: the integration is multi-vendor, so it is not a Palo Alto differentiator by itself.
Evidence and measured results
Opened and read the full Help Center article, including how it works, supported vendors, FAQ and data-handling notes.
Limitations and uncertainty
Beta feature documentation; describes mechanism, not detection effectiveness. Vendor documentation from Anthropic, not an independent evaluation.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-03; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Agencies need to know exactly what desktop-agent monitoring covers before approving Claude Cowork.
- Stakeholders
- AI platform administrators
- Endpoint teams
- Privacy and records officers
- CISOs
- Discovery questions
- Will cloud-run Cowork sessions be allowed?
- Who approves that prompts and files reach the EDR vendor?
- Which Claude Desktop deployment paths are in use?
- Who maintains re-signed packages?
- Value hypothesis
- Interpretation: a clear policy for local monitoring and cloud-session limits may let agencies run controlled pilots.
- Potential engagement
- A desktop-agent monitoring policy and configuration review.
- Unsupported claims
- Complete coverage of all Cowork sessions
- Detection effectiveness
- Palo Alto exclusivity
Pre-sales engineering
Role takeaway
- Fit
- Relevant to any SLED organization piloting Claude Team or Enterprise with Cowork.
- Architecture and integration
- Claude Desktop setting names vendor; signed package copied into VM; sensor reports to vendor cloud.
- Prerequisites
- Team or Enterprise plan
- Approved EDR vendor
- Desktop management tooling
- Constraints
- Beta feature
- Local VM only
- macOS and Windows only
- Security
- Signature check covers only the package file; telemetry includes prompts and files.
- Proof of value
- Verify sensor install on VM start
- Confirm cloud sessions are blocked or separately governed
- Review telemetry for sensitive content handling
Delivery
Role takeaway
- Work
- Configure and validate EDR-in-VM for pilot devices with a cloud-session policy.
- Dependencies
- Claude Desktop admin access
- EDR vendor package
- Privacy approval
- Ownership
- AI platform admin owns settings; endpoint lead owns the sensor; privacy officer owns data approval.
- Skills and adoption
- Admins learn combined Claude Desktop and EDR management.
- Governance checkpoints
- Privacy sign-off
- Cloud-session decision
- Re-signing calendar
- Acceptance criteria
- Proposed: sensor verified on 100% of pilot devices
- Proposed: written cloud-session policy approved
- Proposed: telemetry retention rule documented
- Risks
- Unmonitored cloud sessions
- Sensitive data in EDR logs
- Beta behavior changes
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Sensor deployment is controlled through Claude Desktop settings and vendor-signed packages; data egress from the VM bypasses the Cowork network allowlist.
Governance
Who approves, reviews and stays accountable for outcomes?
Agencies must decide whether cloud-run Cowork sessions are allowed, since they cannot be monitored this way, and must document that agent content flows to the EDR vendor.
Security and privacy
What data, permissions and controls need testing?
Signature checking limits which packages run, but covers only the package file, not later sensor updates. Sensitive prompts and files become EDR telemetry.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Desktop administrators need to manage both Claude Desktop policy and EDR packaging.
Procurement
What should contracts, pricing and exit terms secure?
No extra Anthropic charge during beta; EDR vendor pricing may apply.
Operating model
Which teams own the service once it runs?
Monitoring of desktop agents becomes a joint responsibility of AI platform admins and endpoint teams.
Publication history
- 2026-11-02Palo Alto Networks · Issue 045 resources
Stable resource ID: anthropic-claude-help-center-edr-sensor-in-cowork-vm-beta-local-only-20261102