Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
CVE-2026-125399 — Write-up and proof-of-concept for CVE-2026-125399, a zero-click unauthenticated OS command injection in Fluxion via eval() on attacker-controlled 802.11 ESSID. | Kitploit
Tools/GitHubGitHub/vulnsvnc/cve-2026-125399
Wi-Fi AuditingVulnerability AnalysisExploitationWireless SecurityPapers & ResearchRed Teaming
GitHubvulnsvnc/cve-2026-125399

CVE-2026-125399

Write-up and proof-of-concept for CVE-2026-125399, a zero-click unauthenticated OS command injection in Fluxion via eval() on attacker-controlled 802.11 ESSID.

View Repository
11 day agoNot yet reviewed
Share
Website

CVE-2026-125399 — Zero-Click OS Command Injection PoC & Write-up (Fluxion)

Note for English-speaking readers: This document starts in Turkish. A complete English translation of the entire write-up is included at the bottom of this file, under the heading "ENGLISH VERSION".

ProductFluxion (FluxionNetwork/fluxion)
Tested Versionv6.32 (also present on GitHub master, internal version string is FLUXION V6.38)
Vulnerability ClassCWE-78 (OS Command Injection) via CWE-95 (Eval Injection), root cause CWE-20 (Improper Input Validation of untrusted 802.11 data)
Affected Componentfluxion.sh → fluxion_get_target(), lines ~1839–1844
Attack VectorAdjacent (802.11 / Wi-Fi radio range)
Privilege RequirementNone (on the attacker side)
User InteractionNone (the victim is only performing a routine Wi-Fi scan)
Obtained PrivilegePrivileges of the Fluxion process — usually root (required for monitor mode)
CVSS 3.1AV:A/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H → 8.8 (High), approaching Critical in practice due to running as root
CVECVE-2026-125399
StatusPatched (v6.39)
Discovery DateNovember 4, 2026
Vendor Notification DateNovember 4, 2026

1. Summary

Fluxion parses the list of scanned Wi-Fi networks from airodump-ng output as CSV. To correctly display possible non-ASCII bytes in the SSID field, the code passes the SSID value to Bash's eval builtin after a single and incomplete sanitization step (escaping embedded single quotes). Because this sanitization does not account for a backslash character immediately preceding a single quote within the SSID, an attacker can broadcast a crafted SSID to break out of the quoting context and execute arbitrary shell commands via eval.

The critical point: this code path runs unconditionally for every scanned access point, before the target selection list is shown to the operator. No attack mode needs to be selected and no target needs to be chosen — simply scanning nearby networks is enough. Since Fluxion usually runs as root for monitor mode, the injected command executes with root privileges.

2. Vulnerability Details

2.1 Classification

  • CWE-78 – OS Command Injection
  • CWE-95 – Eval Injection (proximate root cause)
  • CWE-20 – Improper Input Validation (802.11 SSID information element, fully attacker-controlled)

2.2 CVSS v3.1 Vector Detail

MetricValueRationale
AVAdjacentOnly Wi-Fi radio range is required, no network authentication/association needed
ACLowPayload is a static, reliably craftable SSID string; no race condition/timing dependency
PRNoneNo credentials required
UINoneNo interaction required on the victim side beyond a routine scan
SUnchanged
C/I/AHigh/High/HighArbitrary command execution on the operator's machine, usually as root

Base Score: 8.8 (High)

2.3 Root Cause — Vulnerable Code

fluxion.sh, fluxion_get_target() function, approximately lines 1839–1844:

# Parse any non-ascii characters by letting bash handle them.
# Escape all single quotes in ESSID and let bash's $'...' handle it.
local sanitizedESSID=$(
  echo "${candidateAPInfo//\'/\\\'}" | cut -d , -f 14
)
candidatesESSID[i]=$(eval "echo \$'$sanitizedESSID'")

candidateAPInfo, a CSV line produced by airodump-ng; the 14th field is ESSID. According to the 802.11 standard, this field is a free-form byte array of up to 32 bytes that is fully controlled by the attacker. The only sanitization applied is replacing every single quote (') in the SSID with an escaped single quote (\'); after that, the result is embedded in a Bash ANSI-C-quoted string ($'...') and passed to eval.

2.4 Sanitization Bypass Mechanism

The ${candidateAPInfo//\'/\\\'} substitution only rewrites quote characters that already exist in the SSID; it does not account for a backslash character that the attacker places immediately before one of these quotes. If the SSID begins with the bytes \ (backslash) and ' (single quote), the substitution turns this pair into three characters: \\' (backslash, backslash, quote).

When this is evaluated inside Bash's $'...' ANSI-C quoting: the \\ sequence is a recognized escape and is decoded to a single literal backslash, consumed as a unit. The quote character that immediately follows is now unescaped and is interpreted as the actual, closing delimiter of the $'...' string. Everything after this quote is parsed and executed by eval as ordinary shell syntax.

2.5 PoC Payload (Harmless — for proof purposes)

An SSID consisting of the following literal bytes (21 bytes, below the 802.11 limit of 32 bytes) is sufficient:

\'; touch /tmp/pwned #
Byte(s)Purpose
\A literal backslash immediately before the quote to bypass the sanitizer
'When combined with the escaped quote produced by the sanitizer, it closes the $'...' string early
;A shell separator that starts a new command within the line being eval'd
touch /tmp/pwnedAn arbitrary, attacker-chosen command (the PoC payload)
#Turns the rest of the original line into a comment, ensuring it fails safely

The final string reaching eval takes the form: eval "echo $'...\\'; touch /tmp/pwned #...'" — Bash parses this as follows: print the (truncated) SSID text, separately execute the touch /tmp/pwned command, and treat the rest of the original line as a comment.

3. Zero-Click Characteristic and Payload Concealment

3.1 This Is a Zero-Click Vulnerability

What makes this vulnerability particularly severe is that it requires no conscious interaction on the victim's side:

  • The victim does not have to see, select, target, or interact in any way with the attacker's access point.
  • All that is required is to run Fluxion for its normal intended use and perform a routine Wi-Fi scan — an ordinary and unavoidable step that anyone using the tool would already take.
  • The injection is triggered automatically at the moment the attacker's AP is merely physically present among the scanned networks, before any selection screen is shown to the operator.
  • It therefore fits the classic "zero-click" definition exactly: the attacker does not need to request any action (click, confirmation, link, authentication) from the victim to execute the payload; the victim simply using the tool normally is sufficient on its own.

3.2 The Payload Is Structurally Hidden from the Victim

The payload's "concealment" stems not from the attacker visually camouflaging the SSID, but from the architecture of the vulnerable code path:

  • The vulnerable eval call executes while the fluxion_get_target() function builds the candidate network list, that is, before this list is displayed on screen to the operator (see Sections 2.3–2.4 and Section 4).
  • As a result, there is never a moment when the victim could notice the suspicious SSID and make a decision, even after the command has already executed. The opportunity to inspect/suspect is structurally eliminated.
  • Furthermore, within the 32-byte limit permitted by the 802.11 SSID field, the attacker can shorten the visible/printable portion of the payload or combine it with non-printable control characters, so that even if the network list is later displayed in any way (e.g., in logs, on a different screen), the SSID appears as an ordinary/garbled character string. This also makes detection harder retroactively.
  • In summary: what "hides" the payload is not an encryption or obfuscation technique, but the vulnerable code executing the payload before any opportunity for visual inspection — which makes it far more dangerous than typical attacks that require "deceiving the user."
Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    938
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    556
  • 3

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    352
  • 4

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    343
  • 5

    looking-glass

    anonghosty/looking-glass

    CSIRT platform that ingests Shadowserver reports into MongoDB and replays them as interactive maps, analysis, findings, advisories and executive dashboards.

    336
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    291

Explore all tools

Browse our collection of tools

View all tools →