From the Palo Alto Networks edition of November 2, 2026
Cortex Cloud 2.3 adds an eBPF cloud workload sensor, an MCP server for AI coding agents, supply-chain attestation and AI-agent discovery across AWS and GCP
Palo Alto Networks (Cortex Cloud) · Cloud-native application protection (CNAPP) and AppSec · Global product release
- Publisher
- Palo Alto Networks Blog
- Original publication
- Palo Alto Networks blog, Nov. 2, 2026 (Sarit Tager, Cameron Hyde)
- Source retrieved
- 2026-11-03
What happened
Cortex Cloud 2.3 adds Security Initiatives for goal-based remediation tracking; folds in the Cortex Cloud SaaS Security launch (100 SaaS applications); a new eBPF-based Cortex Cloud Sensor for Linux workloads and Kubernetes that the vendor says uses up to 75% fewer resources; an MCP server that brings security context and code scanning to AI coding agents, with native Claude Code and Cursor integrations; a supply-chain dashboard and artifact attestation; open-source reachability analysis; a unified identity dashboard covering human, non-human and AI-agent identities; AI security visibility for AWS Bedrock AgentCore, SageMaker and GCP Gemini Enterprise Agent Platform; IBM Cloud posture support; agentless file integrity monitoring; MCP server discovery; and GitHub Advanced Security integration. It cites a Unit 42 simulated AI-assisted attack reaching exfiltration in 25 minutes. No customer outcomes are reported.
Why it matters
Relevant to state IT agencies, university research computing and large districts that build applications in public cloud and are adopting AI coding assistants. MCP server discovery and AI-agent identity inventory map to emerging state AI-use inventories. The SaaS Security component was archived on Oct. 31; what is new is the workload sensor, coding-agent integration and supply-chain attestation. Limited relevance for small entities without in-house development.
Evidence and measured results
Opened and read the full Nov. 2 blog post. Vendor material only; the 75% resource figure and 25-minute simulation are vendor statements without published methods.
Limitations and uncertainty
Vendor announcement; the 75% resource reduction lacks a baseline or method, and the 25-minute exfiltration figure is from a simulated scenario. Government-cloud availability not stated.
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
- Public entities building cloud applications with AI coding assistants lack visibility into agent-generated code and build trust.
- Stakeholders
- State IT application leads
- University research computing directors
- DevSecOps teams
- CISOs
- Discovery questions
- Which teams use AI coding agents today?
- Which Linux and Kubernetes environments would host the sensor?
- Is artifact attestation required by any state or federal grant terms?
- Is the platform available in the required cloud environment?
- Value hypothesis
- Interpretation: integrating security checks into coding-agent workflows may catch weaknesses earlier for teams already using those agents.
- Potential engagement
- A single-pipeline pilot of the MCP integration, eBPF sensor and attestation.
- Unsupported claims
- The 75% resource reduction in the customer environment
- Any measured vulnerability reduction
- Government-cloud availability
Pre-sales engineering
Role takeaway
- Fit
- Fits SLED organizations with in-house cloud development; limited fit for entities consuming only SaaS.
- Architecture and integration
- eBPF sensor plus agentless scanning; MCP server for Claude Code and Cursor; SBOM and attestation in CI; identity dashboard spanning AI agents.
- Prerequisites
- Supported kernel versions
- Approved AI coding-tool policy
- CI pipeline access
- Cloud account onboarding
- Constraints
- Government-cloud availability not stated
- Kernel compatibility
- Developer tool approvals vary by agency
- Security
- Inventory and authenticate MCP servers; scope coding-agent repository access; enforce attestation before production deploys.
- Proof of value
- Measure sensor CPU and memory on representative nodes
- Confirm MCP server rejects unauthenticated agent calls
- Block an unattested artifact in a test deploy
Delivery
Role takeaway
- Work
- Stage the sensor, MCP integration and attestation through a pilot pipeline.
- Dependencies
- Platform team
- Developer group
- CI owners
- Ownership
- The cloud security lead owns the sensor; application owners own pipeline enforcement.
- Skills and adoption
- Developers need guidance on acting on agent findings; operators need eBPF troubleshooting skills.
- Governance checkpoints
- AI coding-tool policy approval
- Pilot overhead review
- Attestation enforcement decision
- Acceptance criteria
- Proposed: measured sensor overhead documented per node type
- Proposed: 100% of pilot pipeline artifacts attested
- Proposed: all MCP servers in dev environments inventoried
- Risks
- Unscoped agent repository access
- Sensor incompatibility on legacy hosts
- Overreliance on agent-generated fixes
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?
The eBPF sensor adds a kernel-level agent on Linux workloads alongside agentless scanning; the MCP server connects security tooling to developer AI agents, adding a new integration path into CI and IDE environments.
Governance
Who approves, reviews and stays accountable for outcomes?
AI coding agents calling a security MCP server should be covered by agency AI-use and software-assurance policies, including who approves agent access to code repositories.
Security and privacy
What data, permissions and controls need testing?
Artifact attestation and SBOM-based trust support software-supply-chain controls; MCP servers themselves are an attack surface and should be inventoried, which the release also offers.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Developers must learn to act on agent-surfaced findings; security teams need eBPF and Kubernetes operating skills.
Procurement
What should contracts, pricing and exit terms secure?
Confirm whether the sensor, MCP server and SaaS Security require separate licensing, and whether the platform is authorized for the agency's compliance environment.
Operating model
Which teams own the service once it runs?
Security Initiatives frame remediation as tracked goals, which suits agencies reporting risk reduction to oversight bodies.
Publication history
- 2026-11-02Palo Alto Networks · Issue 045 resources
Stable resource ID: pan-cortex-cloud-2-3-ebpf-sensor-mcp-server-ai-coding-agents-supply-chain-20261102