Vulnerability Overview
CVE-2026-35616 is a critical improper access control vulnerability (CWE-284) in Fortinet FortiClientEMS, the centralized endpoint management server used to deploy and manage FortiClient agents across enterprise environments. The flaw allows an unauthenticated, remote attacker to execute unauthorized code or commands by sending crafted requests over the network. With a CVSS base score of 9.8 and a network-based attack vector requiring no privileges or user interaction, this vulnerability represents a severe risk to any organization exposing FortiClientEMS to reachable network segments.
CISA has cataloged this CVE in the Known Exploited Vulnerabilities (KEV) list, confirming active exploitation in the wild.
Affected Product and Severity
| Field | Detail |
|---|---|
| Vendor | Fortinet |
| Product | FortiClientEMS |
| Affected Versions | 7.4.5 through 7.4.6 |
| CVSS 3.x Base Score | 9.8 (Critical) |
| Attack Vector | Network |
| Authentication Required | None |
| CWE | CWE-284 — Improper Access Control |
| KEV Status | Listed by CISA |
Request Path and Exploitation Flow
The following diagram illustrates the attack surface and where compensating controls may intercept malicious traffic:
Attacker Network Edge FortiClientEMS (7.4.5–7.4.6)
│ │ │
│── Crafted HTTP(S) ──────►│ │
│ Request │── [F5 XC WAAP] ──────────────►│
│ │ WAF Policy Evaluation │
│ │ IP / Rate Limiting │
│ │ Protocol Validation │
│ │ │
│ │ IF request passes ──────────►│── Improper Access
│ │ │ Control Bypass
│ │ │── Unauthorized Code
│ │ │ Execution
Because the vulnerability is triggered via crafted network requests, any HTTP/HTTPS-layer proxy or inspection point positioned in front of FortiClientEMS has an opportunity—though not a guarantee—to detect or disrupt exploit traffic before it reaches the vulnerable service.
Priority Action: Patch the Affected Product
Fortinet has published advisory FG-IR-26-099 addressing this vulnerability. Organizations running FortiClientEMS 7.4.5 or 7.4.6 should:
- [ ] Immediately consult the Fortinet advisory and upgrade to a patched release.
- [ ] If patching cannot be performed immediately, restrict network access to the FortiClientEMS management interface to trusted administrative subnets only.
- [ ] Review FortiClientEMS logs for indicators of compromise, particularly anomalous unauthenticated requests.
- [ ] Follow any CISA BOD 22-01 remediation timelines applicable to your organization.
No compensating control replaces applying the vendor-supplied patch.
Compensating Controls: F5 Distributed Cloud WAAP (XC)
For organizations in North America that proxy FortiClientEMS management traffic through F5 Distributed Cloud WAAP, the following capabilities may help reduce exposure while patching is underway. These are generic, defense-in-depth measures—no specific managed WAF rule or verified CVE-specific signature coverage for CVE-2026-35616 has been confirmed in the supplied evidence.
WAF Service Policies. Custom WAF policies can be configured to inspect inbound requests destined for FortiClientEMS endpoints. Policies targeting unusual request patterns, oversized headers, or unexpected HTTP methods may disrupt certain exploit payloads.
HTTP Protocol Validation. Strict HTTP protocol compliance enforcement can reject malformed or non-standard requests that deviate from RFC specifications, which may interfere with crafted exploit traffic.
IP-Based Access Controls. Namespace-level or load-balancer-level IP allowlists can restrict access to FortiClientEMS to known administrative source ranges, directly reducing the unauthenticated attack surface.
Rate Limiting. Applying rate-limiting rules to FortiClientEMS-bound traffic can slow automated exploitation attempts and provide additional time for detection and response.
All configurations should be validated in a non-production environment before deployment. The effectiveness of these controls against CVE-2026-35616 specifically has not been independently verified, and the precise structure of the crafted requests used in exploitation has not been publicly detailed at the time of writing.
Limitations
Because the exact exploit payload structure is not publicly documented, HTTP-layer controls may not fully prevent exploitation. If FortiClientEMS uses non-HTTP protocols or proprietary binary channels for the vulnerable functionality, edge-layer WAF inspection may not be applicable to all attack paths.