← Back to Feed CACHED · 2026-05-17 09:42:19 · cache_key CVE-2025-29912
CVE-2026-0113 · CWE-787 · Disclosed 2026-03-10

In ns_GetUserData of ns_SmscbUtilities

ASSESSED — NOISGATE V0.5
Vendor
Reassessed
Verdict:
01 · The Real Story

This is less an open front door than a landmine hidden in the carrier side of the road

CVE-2026-0113 is an out-of-bounds write in ns_GetUserData inside ns_SmscbUtilities.c, a cellular-modem parsing path tied to SMS Cell Broadcast handling. The CNA text says it can cause remote elevation of privilege with no user interaction; Google's public records only pin the fix to Pixel-family devices at security patch level 2026-03-05 or later, and the Android OSV entry marks the affected range as Pixel-family specific before 2026-03-05.

Google publicly labels the issue High in the March 2026 Pixel bulletin, but for enterprise patch triage that overstates the day-to-day risk. The decisive friction is attacker position: this is not an internet-exposed web service or email client bug, it is a modem-side parser issue that likely requires carrier-path access or a rogue base-station style delivery setup to feed malicious cell-broadcast traffic into the target. That keeps it dangerous but not fleet-fire-drill dangerous.

"= ASSESSED AT MEDIUM: scary bug class, but the attacker needs cellular delivery-path control, not a normal internet foothold"
02 · The Attack Path

4 steps from start to impact.

STEP 01

Obtain cellular delivery-path control

The attacker first needs a way to inject or relay crafted cell-broadcast/SMSCB traffic toward the target handset. In practice that means either telecom-side access to broadcast infrastructure or a specialized rogue-radio setup using SDR and open-source RAN tooling such as srsRAN; academic baseband work shows modem parsers are reachable from rogue base stations.
Conditions required:
  • RF proximity to the victim or access to carrier/core broadcast infrastructure
  • Specialized radio equipment and protocol knowledge, or telecom insider/partner access
  • Target device must have the affected Google cellular modem stack
Where this breaks in practice:
  • This is not reachable from the public internet like a VPN or web-admin bug
  • Rogue base-station operations are technically noisy, regulated, and operationally risky
  • Only the relevant Pixel device population is publicly confirmed
Detection/coverage: Traditional network scanners, EDR, and ASM platforms have almost no direct visibility into this prerequisite. Detection is mostly indirect through RF monitoring, carrier telemetry, or device patch inventory.
STEP 02

Deliver malformed SMSCB payload into ns_GetUserData

The crafted broadcast payload must survive radio/protocol handling and hit the vulnerable bounds check in ns_GetUserData within ns_SmscbUtilities.c. Because the bug sits in a parser, exploitation likely depends on precise message structure and modem/firmware behavior rather than generic packet spray.
Conditions required:
  • Ability to send protocol-valid but malicious cell-broadcast content
  • Target modem firmware version predates the 2026-03-05 fix
  • The vulnerable code path must be enabled in that modem build
Where this breaks in practice:
  • Public technical details are sparse and the bug ID is not publicly opened
  • Baseband exploitation is fragile across modem variants and firmware revisions
  • Malformed traffic may be normalized or dropped before reaching the exact parser state
Detection/coverage: No mainstream vuln scanner validates this parser path. At best, defenders can infer exposure from model inventory plus security patch level.
STEP 03

Convert memory corruption into modem-side code execution or privilege gain

An out-of-bounds write gives the attacker a memory-corruption primitive, but turning that into reliable execution or privilege escalation inside a modern modem stack still requires exploit engineering against mitigations and vendor-specific layouts. Google's classification says EoP, which suggests privileged impact exists, but the public record does not show the exact post-corruption primitive.
Conditions required:
  • Reliable control over the overwritten memory region
  • A modem build where exploit mitigations can be bypassed
  • A stable target-specific exploit chain
Where this breaks in practice:
  • Memory corruption in baseband firmware is harder to weaponize than the CNA text alone implies
  • Lack of public PoC suggests the chain is non-trivial or at least not yet commoditized
  • Operational reliability matters more here than in a crash-only bug
Detection/coverage: Most endpoint tools do not instrument modem firmware internals. Crashes may show up as radio instability or baseband resets, but exploit success can remain largely opaque.
STEP 04

Reach meaningful device impact

To matter to enterprise defenders, the attacker needs more than a transient modem crash; they need persistence, surveillance capability, or a path to broader device privilege. That is plausible in a baseband bug class, but the public advisory does not show a demonstrated cross-boundary chain from modem compromise to repeatable full-device takeover across the fleet.
Conditions required:
  • Successful post-corruption control in the modem environment
  • A usable follow-on objective such as interception, surveillance, or pivot into higher-privilege device services
  • A victim population worth the specialized delivery effort
Where this breaks in practice:
  • Blast radius is mostly per-device, not one-shot tenant-wide compromise
  • Enterprise mobile fleets are usually a smaller subset than server/workstation estates
  • No public exploitation evidence or KEV inclusion raises the bar for emergency treatment
Detection/coverage: Mobile EDR/MDM can confirm patch state and sometimes radio anomalies, but it usually cannot prove exploitation of a modem parser bug.
03 · Intelligence Metadata

The supporting signals.

In-the-wild statusNo CISA KEV listing and no authoritative public reporting of active exploitation was found in the sources reviewed here.
Proof-of-concept availabilityNo public CVE-specific exploit or GitHub PoC was found. Realistic delivery would likely involve carrier/CBC access or rogue-RAN tooling such as srsRAN.
EPSSPrompt-supplied EPSS = 0.00238 (~0.238% 30-day exploitation probability), which is low and consistent with a specialized attack path.
KEV statusNot listed in the CISA Known Exploited Vulnerabilities Catalog.
Technical classMemory corruption via out-of-bounds write in a cellular modem parser; third-party databases map it to CWE-787. No official CVSS vector or score was published.
Affected versionsPublicly, Google exposes this as a Pixel bulletin issue in the Cellular Modem subcomponent. OSV marks the affected range as Pixel-family specific before security patch level 2026-03-05.
Fixed versionsGoogle states security patch level 2026-03-05 or later addresses the issue for supported Pixel devices.
Scanning / exposure data*Inference from component location:* this is a device-side modem/RF parser, so there is no meaningful Shodan/Censys/FOFA-style internet exposure count to prioritize against. Your exposure data comes from mobile inventory + patch level, not ASM.
Disclosure timelineCVE record published 2026-03-10 by Google_Devices; Pixel bulletin published 2026-03-03 and updated 2026-03-09.
Researcher / reporting orgPublic credit was not exposed in the sources reviewed. The issuing CNA is Google_Devices.
04 · The Call

noisgate verdict.

Final Verdict
DOWNGRADED to MEDIUM (5.9/10)

This lands at MEDIUM because the single biggest limiter is attacker position: exploitation likely requires cellular delivery-path control rather than ordinary internet reachability. That sharply reduces the exposed population and prevents the bug from behaving like a normal remotely sprayable enterprise vulnerability, even though the underlying bug class is serious.

MEDIUM overall severity assessment
HIGH patch-level and affected-population facts from Google bulletins/OSV
LOW exact exploit-chain reliability because the bug details are not public

Why this verdict

  • Downward pressure: attacker must control the cellular delivery path rather than just reach an exposed IP service. That implies RF proximity, carrier-side access, or telecom-grade equipment and expertise.
  • Downward pressure: reachable population is narrower than the CVE text suggests. Publicly confirmed exposure is tied to supported Pixel-family devices before 2026-03-05, not a broad cross-vendor Android server estate.
  • Upward pressure: the bug is still pre-auth remote memory corruption with no UI in a privileged modem parsing path, so it cannot be dismissed as harmless or purely theoretical.

Why not higher?

There is no public exploitation evidence, no KEV inclusion, and no public PoC. More importantly, the path is not internet-scalable: most attackers cannot casually inject malicious cell-broadcast traffic at enterprise targets the way they can spray HTTP, VPN, or email attack surfaces.

Why not lower?

The CNA impact statement is still serious: remote, no user interaction, and memory corruption in a modem-facing parser. If you run a managed Pixel fleet for executives, journalists, field teams, or travelers, the threat model is materially stronger than a routine nuisance bug.

05 · Compensating Control

What to do — in priority order.

  1. Enforce Pixel patch compliance — Use MDM/UEM attestation and compliance policies to identify Google devices below security patch level 2026-03-05 and block them from sensitive access until updated. For a MEDIUM verdict there is no mitigation SLA; use this as an interim control while you work inside the 365-day remediation window.
  2. Segment mobile access by trust level — Require stronger conditional access for outdated mobile devices, especially for email, chat, and high-value SaaS sessions. This does not fix the modem bug, but it reduces business impact if a vulnerable handset is compromised before you complete patching.
  3. Prefer managed travel profiles for high-risk staff — For executives, journalists, and staff traveling to hostile environments, use the newest patched devices and stricter network/device-trust policies. This is where a specialized RF-path bug matters most, even though there is no noisgate mitigation SLA for a MEDIUM item.
  4. Monitor mobile patch drift continuously — Build a standing report for manufacturer=Google and security_patch < 2026-03-05 so this does not disappear into the general Android noise floor. The real control here is inventory discipline, because standard vulnerability scanners will not see this parser directly.
What doesn't work
  • A WAF does nothing here because the attack surface is not HTTP.
  • Email filtering does nothing because the delivery path is cellular broadcast/modem-side, not mail-borne content.
  • Mobile app allowlisting is not a fix because the vulnerable code path is in modem firmware/system components below the app layer.
  • Internet-facing ASM scanning will not tell you who is vulnerable; this is an RF/device exposure problem, not a public-IP exposure problem.
06 · Verification

Crowdsourced verification payload.

Run this on an auditor workstation with the Android SDK adb tool installed, pointed at a USB-connected or network-connected device with debugging enabled. Invoke it as ./check_cve_2026_0113.sh <device-serial> or omit the serial for a single attached device; no root is required.

noisgate-verify.sh
BASHREAD-ONLYSAFE
#!/usr/bin/env bash
# check_cve_2026_0113.sh
# Verify whether a connected Google Pixel device is patched for CVE-2026-0113
# Logic: Google publicly fixes this issue at security patch level 2026-03-05 or later.
# Exit codes: 0=PATCHED, 1=VULNERABLE, 2=UNKNOWN, 3=usage/adb error

set -euo pipefail

TARGET_PATCH="2026-03-05"
SERIAL="${1:-}"
ADB=(adb)

if ! command -v adb >/dev/null 2>&1; then
  echo "UNKNOWN: adb not found in PATH"
  exit 3
fi

if [[ -n "$SERIAL" ]]; then
  ADB+=( -s "$SERIAL" )
fi

# Verify connectivity
if ! "${ADB[@]}" get-state >/dev/null 2>&1; then
  echo "UNKNOWN: no reachable adb device"
  exit 2
fi

getprop_safe() {
  local key="$1"
  "${ADB[@]}" shell getprop "$key" 2>/dev/null | tr -d '\r'
}

MANUFACTURER="$(getprop_safe ro.product.manufacturer)"
BRAND="$(getprop_safe ro.product.brand)"
MODEL="$(getprop_safe ro.product.model)"
PATCH="$(getprop_safe ro.build.version.security_patch)"
BUILD="$(getprop_safe ro.build.fingerprint)"

if [[ -z "$PATCH" ]]; then
  echo "UNKNOWN: could not read ro.build.version.security_patch"
  exit 2
fi

# Normalize vendor check for Google Pixel-family devices.
shopt -s nocasematch
IS_GOOGLE=0
if [[ "$MANUFACTURER" == *google* ]] || [[ "$BRAND" == *google* ]] || [[ "$MODEL" == Pixel* ]]; then
  IS_GOOGLE=1
fi
shopt -u nocasematch

if [[ "$IS_GOOGLE" -ne 1 ]]; then
  echo "UNKNOWN: device is not clearly a Google Pixel device"
  echo "manufacturer=$MANUFACTURER brand=$BRAND model=$MODEL patch=$PATCH"
  exit 2
fi

# YYYY-MM-DD sorts lexicographically.
if [[ "$PATCH" < "$TARGET_PATCH" ]]; then
  echo "VULNERABLE: Google device patch level is below $TARGET_PATCH"
  echo "manufacturer=$MANUFACTURER model=$MODEL patch=$PATCH"
  echo "build=$BUILD"
  exit 1
else
  echo "PATCHED: Google device patch level is $PATCH (>= $TARGET_PATCH)"
  echo "manufacturer=$MANUFACTURER model=$MODEL"
  echo "build=$BUILD"
  exit 0
fi
07 · Bottom Line

If you remember one thing.

TL;DR
Monday morning, pull a clean inventory of Google/Pixel devices and identify anything below security patch level 2026-03-05. Because this is MEDIUM, there is no noisgate mitigation SLA — go straight to the 365-day remediation window; use conditional-access and patch-compliance gates as sensible interim controls, but the real goal is to get vulnerable managed Pixels updated within the noisgate remediation SLA of ≤ 365 days. If you have high-risk travelers or executive targets, front-load them anyway even though this is not a general fleet emergency.

Sources

  1. Google Pixel Update Bulletin — March 2026
  2. Android Security Bulletin — March 2026
  3. Android OSV entry PUB-A-441209133
  4. OpenCVE record for CVE-2026-0113
  5. CISA Known Exploited Vulnerabilities Catalog
  6. srsRAN Project documentation
  7. BaseSAFE: Baseband Sanitized Fuzzing through Emulation
  8. Google Pixel Help — Check & update your Android version
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.