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-2026-19971_guest — LB-LINK WR1210M (CVE-2026-19971) 未授权配置泄露漏洞与访客网络扩展攻击链分析 | Kitploit
Tools/GitHubGitHub/tianlansedexiaoniu/cve-2026-19971_guest
Embedded Systems SecurityIoT SecurityVulnerability AnalysisWeb Application ExploitationNetwork SecurityWireless SecurityPapers & Research
GitHub
Share
tianlansedexiaoniu/cve-2026-19971_guest

CVE-2026-19971_guest

LB-LINK WR1210M (CVE-2026-19971) 未授权配置泄露漏洞与访客网络扩展攻击链分析

View Repository
13 days agoNot yet reviewed

LB-LINK (Blink) Brand Router Guest Network Firewall Isolation Bypass Attack Chain Analysis Report

Extended exploit chain based on CVE-2026-19971 (backup.cgi unauthorized configuration disclosure) · First disclosed by the author · Verified on real hardware


The Original CVE in One Sentence

CVE-2026-19971 (LB-LINK brand devices, /www/cgi-bin/backup.cgi missing authentication, CWE-306 / CWE-287) was disclosed by m1nkvpm on 2026-08-17 via VulDB (VDB-391148, submission #873139). Its core description covers "unauthorized configuration retrieval from the trusted domain (LAN / primary WiFi)" — see the original CVE for details; this report does not elaborate further.


0x01 Discovery Statement for This Attack Chain

  • Discovery and disclosure: This "guest network → firewall isolation bypass → reach backup.cgi" attack chain was independently discovered and first publicly disclosed by the author of this report, and has been verified on real hardware (a certain router model under the LB-LINK brand, which may or may not be the same model as the one in the original CVE disclosure).
  • Relationship to the base CVE: The fatality of the base CVE is confined within the trusted network by "firewall zone isolation" — the original disclosure assumes that the guest network, due to isolation, cannot reach backup.cgi. This chain proves that even when guest isolation is enabled in the configuration, a transient window still exists that allows it to be bypassed, and the attack surface spills over from the "internal network" to "any stranger connected to the guest WiFi".

0x02 Vulnerability Mechanism (Core)

Mechanism in one sentence: The router enables firewall rules for the guest WiFi that isolate it from the main network, but at the instant the isolation / firewall service reloads or the interface reconnects, the ruleset has an ms-level gap — at this moment the isolation has not yet been fully issued / applied. An attacker within the guest network actively and repeatedly triggers enable ↔ disable of that service, forcing the rules to reload repeatedly, and within each reload's transient window initiates access to backup.cgi and captures the response packets, thereby obtaining a complete configuration backup while "guest isolation is nominally still in effect".

Three key points:

  1. Source of the window (public corroboration): During firewall rule reloads, the ruleset being briefly empty / incrementally issued is a known race window in the OpenWrt ecosystem. OpenWrt's official firewall4 "drop input by default" commit states: at boot / reload the ruleset is mostly empty and interface rules are added incrementally, "if a packet hits before the isolation / DNAT rules are in place, it will cause incorrect forwarding or bypass"; the NAT masquerade reload race mailing list thread also recorded "a transient window where filter rules are not in place and nat rules are missing". This brand's devices (firmware system based on OpenWrt) inherit this ecosystem, and therefore also have this window.
  2. The window can be induced: Unlike the base CVE's "passively waiting for a natural reconnect", this chain actively creates reload events by repeatedly toggling the guest isolation service, without relying on the device reconnecting naturally — the attacker can control the timing themselves.
  3. Depends on a timing window: Yes. The bypass depends on the transient window (ms-level) of the WiFi / service reconnect, requiring precise timing to capture; it is not "reachable at any time".

0x03 Impact and Changes to the Attack Surface

DimensionBase CVE (original disclosure)This attack chain (guest bypass)
Reachable domainTrusted LAN / primary WiFiUntrusted guest WiFi
Attacker profileInternal / semi-trusted userAny stranger connected to the guest WiFi
Trigger methodDirect connection from trusted domain / waiting for reconnect windowActively induced reload window
Exposure radiusInside the device owner's networkExpanded to public guest spaces (stores, cafes, etc.)
  • Once the backup.cgi response is obtained within the window, its 32-byte private header can be stripped, the package unpacked to extract the etc/* configuration tree, and all credentials recovered — Web admin password, WiFi PSK, root (SSH) hash, etc. — fully consistent with the credential leakage of the base CVE, but because the attacker is less controllable, the consequences are more irreversible.

0x04 Similar Research and Corroboration (Evidence That This Chain Is Not an Isolated Case)

  • AirSnitch (NDSS 2026, Vanhoef et al.): Systematically demonstrates that guest/client isolation can be bypassed on mainstream routers (Netgear / TP-Link / ASUS / OpenWrt, etc.), with attacks able to cross the guest boundary into the main network. This shows that "guest isolation" itself is fragile and non-standardized.
  • CVE-2025-46635 (Tenda RX2 Pro): Guest WiFi network isolation bypass, where an attacker in the guest domain crosses into the trusted subnet via static IP and other means; this belongs to the same class of "guest isolation failure" defects.
  • OpenWrt firewall reload race window: See 0x02 Key Point 1, which provides direct technical evidence for the window mechanism of this chain.

This chain is the concrete instantiation and first public disclosure of the aforementioned "guest isolation bypass" class of defects on LB-LINK brand routers.


0x05 Remediation Recommendations

  1. Apply firewall rules atomically: Maintain "default deny" during reload/reconnect to eliminate ruleset gaps (refer to the OpenWrt firewall4 "drop input by default" remediation approach—accept no inbound/forward traffic until reload is complete).
  2. Prohibit guest-domain-induced reloads: Remove or rate-limit the guest network's ability to trigger isolation/firewall service enable / disable, eliminating the possibility of "actively creating a window."
  3. Close the root cause of the base CVE: Make backup.cgi use session authentication consistent with the main management plane, allowing only logged-in users to download backups.
  4. User side: Avoid opening guest WiFi in public/commercial venues; if opened, ensure isolation rules do not have gaps during reload, and regularly change all passwords.

Disclaimer

This report is intended solely for security research and defensive hardening. It contains no reproducible PoC, EXP, HTTP request samples, Payload, or exploit code; all paths are presented as mechanism and logic descriptions and cannot be directly used for unauthorized testing. It is strictly prohibited to use this document for device testing without explicit authorization. For vulnerability disclosure, please follow responsible disclosure procedures, notify the vendor first, and reserve a remediation window.

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.

    945
  • 2

    ghostlock-app

    yukonga/ghostlock-app

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

    555
  • 3

    redforge

    cybersecurity91/redforge

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

    350
  • 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.

    339
  • 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.

    338
  • 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.

    292

Explore all tools

Browse our collection of tools

View all tools →