← Back to Feed CACHED · 2026-08-13 07:15:47 · CACHE_KEY CVE-2026-18622
CVE-2026-18622 · CWE-451 · Disclosed 2026-08-13

Foxit PDF Editor/Reader's signature-validation pop-up reports modified certified documents as valid

ASSESSED — NOISGATE V0.5
Vendor
Reassessed
Verdict:
Do you agree?
01 · The Real Story

A notary stamp that keeps saying 'authentic' even after someone swaps the pages

CVE-2026-18622 is a UI-misrepresentation flaw (CWE-451) in Foxit PDF Reader and Foxit PDF Editor on Windows and macOS. When a digitally signed or certified PDF has its signature fields abnormally modified — through changes to appearance properties, field coordinates, or outright field duplication — the application's signature-validation pop-up still reports the document as validly signed. This lets an attacker alter visible document content post-signature and present a forgery that Foxit's UI blesses as trustworthy. Affected versions include all Foxit PDF Reader and Editor builds prior to 2026.1.3 on both platforms, plus Editor legacy tracks prior to 14.0.6 and 13.2.6. The researchers — Enzo da Rosa Brum, Frederico Schardong, and Ricardo Felipe Custódio at Brazil's Federal University of Santa Catarina — build on the well-known 'shadow attack' lineage of PDF signature bypasses first cataloged at pdf-insecurity.org.

Foxit's own MEDIUM / 4.7 rating is honest and, if anything, generous. The CVSS vector AV:L/AC:H/PR:N/UI:R/S:U/C:N/I:H/A:N correctly captures that (a) the attack is delivered via a local file the victim must open, (b) complexity is high because the attacker must craft a specific manipulation that survives Foxit's partial validation logic, (c) there is no code execution — the impact is purely integrity (document trust), and (d) user interaction is required. No confidentiality or availability impact exists. This is a social-engineering amplifier for document fraud, not a gateway to system compromise.

"Signature UI lies about tampered PDFs, but only in Foxit, only when users check signatures, and no code runs."
02 · The Attack Path

3 steps from start to impact.

STEP 01

Craft a shadow / field-manipulation PDF

The attacker starts with a legitimately signed or certified PDF and modifies signature field attributes — appearance streams, bounding-box coordinates, or duplicated field dictionaries — so that the visible content no longer matches what was originally signed. The PDF specification's flexibility around incremental updates and annotation overlays provides ample room for this manipulation. Prior academic work (Ruhr-Uni Bochum shadow attacks, 2020–2021) provides public tooling and methodology.
Conditions required:
  • Attacker possesses or intercepts a signed PDF
  • Knowledge of PDF incremental-update and signature-field internals
Where this breaks in practice:
  • AC:H — not every field-manipulation strategy survives Foxit's partial validation; attacker must test against the target Foxit version
  • No public PoC exploit code specific to CVE-2026-18622 is available as of disclosure day
STEP 02

Deliver the crafted PDF to the victim

The attacker sends the manipulated PDF via email, file share, or web download. Because the file is a valid PDF (not an executable), most email gateways and endpoint protection products will not block it. The file must be opened specifically in a vulnerable version of Foxit — if the recipient uses Adobe Acrobat, a browser's built-in viewer, or a patched Foxit build, the attack fails.
Conditions required:
  • Victim uses Foxit PDF Reader or Editor (not Adobe, not browser viewer)
  • Victim's Foxit version is pre-2026.1.3 / pre-14.0.6 / pre-13.2.6
Where this breaks in practice:
  • Foxit holds roughly 14–25% of the PDF reader market; majority of enterprise desktops default to Adobe or browser viewers
  • Email-based delivery may be flagged if the PDF contains suspicious JavaScript, though this CVE does not require JS
Detection/coverage: Foxit version inventory via SCCM, Intune, or vulnerability scanners (Tenable plugin expected shortly). No network-level detection is meaningful — the attack is in the rendered UI, not in network traffic.
STEP 03

Victim inspects signature and trusts the forgery

The victim opens the PDF, sees Foxit's signature-validation banner or pop-up reporting the document as validly signed, and trusts the altered content. This is the exploitation moment — the user makes a business decision (approve a contract, accept an invoice, trust a regulatory filing) based on the false validation. The impact is integrity-only: no shell, no data exfiltration, no persistence.
Conditions required:
  • Victim actually checks the digital signature (many users ignore signature UI entirely)
  • Victim makes a consequential decision based on the signature status
Where this breaks in practice:
  • Many enterprise PDF workflows do not rely on client-side signature validation — server-side validation (e.g., DocuSign, Adobe Sign backend) is common
  • Even when users check signatures, a tampered PDF may have other visual inconsistencies that raise suspicion
Detection/coverage: No runtime detection is possible — the 'exploit' is the UI displaying incorrect information. Post-incident forensic validation with a different PDF viewer (Adobe, pdf-parser.py) would reveal the tampering.
03 · Intelligence Metadata

The supporting signals.

In-the-wild exploitationNone known. Not on CISA KEV. No reports of active campaigns. This is a research-driven disclosure from academia.
Proof-of-conceptNo public PoC specific to CVE-2026-18622 as of disclosure day (2026-08-13). Generic PDF shadow-attack tooling exists from Ruhr-Uni Bochum research (pdf-insecurity.org) but requires adaptation.
EPSS scoreNot yet scored — CVE was published today (2026-08-13). FIRST EPSS typically populates within 24–48 hours. Expected to be low given AC:H and no known exploitation.
KEV statusNot listed. No CISA KEV entry.
CVSS vectorCVSS:3.1/AV:L/AC:H/PR:N/UI:R/S:U/C:N/I:H/A:N — Local vector, high complexity, user interaction required, integrity-only impact. No confidentiality or availability damage.
Affected versionsFoxit PDF Reader < 2026.1.3 (Win/Mac). Foxit PDF Editor < 2026.1.3, < 14.0.6, < 13.2.6 (Win/Mac).
Fixed versionsFoxit PDF Reader 2026.1.3, Foxit PDF Editor 2026.1.3 / 14.0.6 / 13.2.6 (all platforms).
Scanning / exposureClient-side desktop application — not directly observable via Shodan/Censys/GreyNoise. Asset inventory (SCCM, Intune, Tanium, Qualys agent) is the detection path. Foxit holds ~14–25% PDF market share; enterprise penetration skews lower due to Adobe dominance in regulated industries.
Disclosure date2026-08-13 (coordinated disclosure via Foxit security bulletin).
Reporting researchersEnzo da Rosa Brum, Frederico Schardong, Ricardo Felipe Custódio — Federal University of Santa Catarina (UFSC), Brazil. Schardong is a recognized digital-identity researcher (2024 Vittorio Bertocci Awardee).
04 · The Call

noisgate verdict.

Final Verdict
= UNCHANGED to MEDIUM (4.2/10)

The single most decisive factor is blast radius limited to per-document social engineering with no code execution — a successful attack convinces one user to trust one forged document, with no lateral movement, no persistence, and no system compromise. The AC:H rating and the requirement that the victim specifically use Foxit (minority market share in enterprise) compound to keep real-world risk squarely in the MEDIUM bucket.

HIGH Vulnerability mechanics and affected versions (confirmed via Foxit bulletin)
HIGH No active exploitation or public PoC
MEDIUM Enterprise Foxit deployment prevalence (market share data varies 14–25%)

Why this verdict

  • No code execution: This is an integrity-only UI-misrepresentation bug. The worst outcome is document fraud — no shell, no privilege escalation, no data exfiltration. This caps the ceiling well below HIGH.
  • High attack complexity (AC:H): The attacker must craft a PDF manipulation that specifically survives Foxit's partial signature validation logic. Not every shadow-attack variant works; version-specific testing is required.
  • Minority viewer market share: Foxit holds roughly 14–25% of the PDF market. In enterprise environments with Adobe Acrobat site licenses or browser-default viewing, the reachable victim population shrinks further.
  • Role multiplier: Foxit PDF Reader/Editor is a *desktop client application*. It does not occupy any high-value infrastructure role (not an IdP, hypervisor, domain controller, or network edge device). Even in document-signing workflows, server-side validation backends (DocuSign, Adobe Sign) are unaffected. The blast radius is one user, one document — no fleet-scale or domain-scale outcome is reachable through this CVE alone.
  • User interaction dependency: The victim must (a) open the PDF in Foxit specifically, (b) inspect the signature validation UI, and (c) make a consequential decision based on it. Each step is a probabilistic filter that reduces real-world impact.

Why not higher?

There is no code execution, no confidentiality impact, and no availability impact. The attack cannot pivot to system compromise, lateral movement, or data exfiltration. The affected component is a desktop PDF viewer — it does not occupy a high-value infrastructure role whose compromise would cascade to fleet-scale impact. No floor-raising condition from the role multiplier applies.

Why not lower?

The integrity impact is genuinely High per CVSS — if a victim trusts the forged signature, the consequences in legal, financial, or regulatory document workflows can be severe. PDF signature trust is a meaningful security property in government and enterprise contexts, and Foxit is deployed in over 1,200 government and corporate offices worldwide. The academic pedigree of shadow attacks (26 of 28 PDF viewers vulnerable in the original research) suggests the manipulation techniques are well-understood and reproducible.

05 · Compensating Control

What to do — in priority order.

  1. Enforce server-side signature validation for high-value document workflows — Route contracts, procurement approvals, and regulatory filings through a server-side validation service (Adobe Sign, DocuSign, or an internal ETSI-compliant validator) that does not rely on Foxit's client-side UI. This neutralizes the attack entirely. No specific SLA — this is a process control, not a patch.
  2. Deploy Foxit 2026.1.3 / 14.0.6 / 13.2.6 through standard software update channels — No mitigation SLA applies for MEDIUM — go straight to the 365-day noisgate remediation SLA. Bundle with your next quarterly desktop software refresh. Use SCCM, Intune, or Foxit's enterprise admin console for deployment.
  3. Educate document-signing stakeholders about cross-viewer validation — Advise users who rely on PDF digital signatures to cross-check signatures in a second viewer (Adobe Acrobat, or a command-line tool like pdfsig from poppler-utils) before making high-stakes decisions. This is a low-cost awareness measure.
  4. Inventory Foxit installations across the fleet — Use your endpoint management platform (SCCM, Intune, Tanium, Qualys agent) to identify all Foxit PDF Reader and Editor installations and their versions. Prioritize patching for users in legal, finance, and procurement roles who handle signed documents.
What doesn't work
  • Email gateway / attachment scanning — the crafted PDF is a valid, well-formed document. It contains no malware, no macros, and no exploit shellcode. Content-inspection engines will pass it through.
  • EDR / endpoint protection — no malicious code executes. The 'exploit' is the UI displaying incorrect information. No behavioral indicator exists for EDR to trigger on.
  • Network-level IDS/IPS — the attack payload is a structurally valid PDF file. Signature-based network detection has no viable detection surface.
  • PDF JavaScript disablement — this CVE does not require JavaScript execution. Disabling JS in Foxit preferences does not mitigate the signature-field manipulation.
06 · Verification

Crowdsourced verification payload.

Run on each Windows endpoint (or remotely via Invoke-Command) with standard user privileges. Example: powershell -ExecutionPolicy Bypass -File Check-CVE-2026-18622.ps1. Also works for macOS by checking the app bundle version manually — but the script targets Windows where Foxit is most commonly deployed.

noisgate-verify.ps1
POWERSHELLREAD-ONLYSAFE
#!/usr/bin/env powershell
# Check-CVE-2026-18622.ps1
# Checks whether installed Foxit PDF Reader or Editor is vulnerable to CVE-2026-18622
# No admin privileges required. Exit codes: 0=PATCHED, 1=VULNERABLE, 2=UNKNOWN

$ErrorActionPreference = 'SilentlyContinue'
$vulnerable = $false
$found = $false

# Define fixed versions per product track
$fixedVersions = @{
    '2026' = [version]'2026.1.3.0'
    '14'   = [version]'14.0.6.0'
    '13'   = [version]'13.2.6.0'
}

# Search common install locations and registry
$regPaths = @(
    'HKLM:\SOFTWARE\Foxit Software\Foxit PDF Reader',
    'HKLM:\SOFTWARE\Foxit Software\Foxit PDF Editor',
    'HKLM:\SOFTWARE\WOW6432Node\Foxit Software\Foxit PDF Reader',
    'HKLM:\SOFTWARE\WOW6432Node\Foxit Software\Foxit PDF Editor'
)

foreach ($rp in $regPaths) {
    $installPath = (Get-ItemProperty -Path $rp -Name 'InstallPath' -ErrorAction SilentlyContinue).InstallPath
    if (-not $installPath) { continue }
    
    $exes = @('FoxitPDFReader.exe','FoxitPDFEditor.exe','FoxitReader.exe','FoxitPhantomPDF.exe')
    foreach ($exe in $exes) {
        $fullPath = Join-Path $installPath $exe
        if (Test-Path $fullPath) {
            $found = $true
            $ver = (Get-Item $fullPath).VersionInfo.ProductVersion
            if (-not $ver) { continue }
            $parsedVer = [version]($ver -replace '[^0-9.]','')
            $majorTrack = $parsedVer.Major.ToString()
            
            if ($fixedVersions.ContainsKey($majorTrack)) {
                $fixed = $fixedVersions[$majorTrack]
                if ($parsedVer -lt $fixed) {
                    Write-Host "VULNERABLE - $exe version $ver (fixed in $fixed)" 
                    $vulnerable = $true
                } else {
                    Write-Host "PATCHED - $exe version $ver"
                }
            } else {
                # Unknown major version track - flag for review
                Write-Host "UNKNOWN - $exe version $ver (unrecognized version track $majorTrack)"
            }
        }
    }
}

if (-not $found) {
    Write-Host 'UNKNOWN - No Foxit PDF Reader or Editor installation detected'
    exit 2
} elseif ($vulnerable) {
    exit 1
} else {
    exit 0
}
07 · Bottom Line

If you remember one thing.

TL;DR
CVE-2026-18622 is a MEDIUM-severity UI-misrepresentation bug — no code runs, no system is compromised, and the attack only works against the minority of enterprise desktops running Foxit. Monday morning, confirm your Foxit installation inventory via SCCM or Intune and identify users in legal, finance, and procurement who handle digitally signed PDFs — those are your priority cohort. There is no noisgate mitigation SLA for MEDIUM findings; go straight to the 365-day noisgate remediation SLA and bundle the update to Foxit 2026.1.3 (or 14.0.6 / 13.2.6 for legacy tracks) into your next quarterly desktop software refresh cycle. If your organization relies heavily on client-side PDF signature validation for legal or regulatory workflows, consider accelerating the patch or implementing server-side signature validation as a compensating control. No active exploitation or public PoC exists, so there is no urgency to break your normal patching cadence.

Sources

  1. Foxit Security Bulletin — CVE-2026-18622
  2. CWE-451: UI Misrepresentation of Critical Information
  3. PDF Shadow Attacks Research — pdf-insecurity.org
  4. Frederico Schardong — Google Scholar
  5. Foxit PDF Editor Market Share — 6sense
  6. Foxit Patches Multiple Vulnerabilities in 2026.1.2 Update — CyberPress
  7. FIRST EPSS API
Peer Review

What defenders are saying.

Submit a review attribution: handle + country only
0 flags selected · stored anonymously
Validation Results

Crowdsourced verification outputs.

Results submitted by users who ran the verification payload against their environment.