CVE-2023-27098: TP-Link Tapo: hardcoded creds bypass login auth
HIGH CISA: TRACK*TP-Link's Tapo companion app for smart cameras and plugs ships hardcoded credentials in versions up to 2.12.703, letting anyone who extracts the string from the APK authenticate to the app's login panel without a real password. The flaw is network-exploitable with no privileges or user interaction required (CVSS 7.5), but real-world urgency is limited: EPSS sits at just 0.376% (top 70th percentile, not the high end), it is not in CISA KEV, CISA's SSVC rates it TRACK_STAR (monitor, no immediate action), and no public exploit code or Nuclei scanner template exists. This is a consumer IoT/smart-home app issue, not an AI/ML pipeline vulnerability — it appears in this feed under an 'ml_ui' tag despite having nothing to do with model serving, inference, or AI tooling, likely because Tapo cameras separately market AI-based motion/object detection features unrelated to this specific bug. Organizations that allow TP-Link Tapo devices on corporate or guest networks (e.g., as physical security cameras) should confirm the app is updated past 2.12.703 and keep these devices segmented on an isolated IoT VLAN regardless of patch status.
What is the risk?
Low-to-moderate operational risk. Exploitability is high in principle (AV:N/AC:L/PR:N/UI:N), but exploitation requires the attacker to first reverse-engineer the APK to recover the hardcoded credential string, and there is no evidence of active exploitation. EPSS (0.376%) and SSVC (TRACK_STAR) both indicate low real-world exploitation probability, and the only known PoC is a GitHub write-up rather than weaponized tooling. Impact is confidentiality-only at the login-panel level (C:H/I:N/A:N) — this is not a full device-takeover or data-destruction primitive.
How does the attack unfold?
What systems are affected?
| Package | Ecosystem | Vulnerable Range | Patched |
|---|---|---|---|
| n/a | — | — | No patch |
Do you use n/a? You're affected.
How severe is it?
What is the attack surface?
What should I do?
1 step-
Update the Tapo app to the latest release via the official app store — TP-Link's fix removes or rotates the hardcoded credential, though no specific patched version is listed in the CVE record, so confirm against TP-Link's support channel. Until confirmed patched, do not rely on the app's login panel as a security boundary, and keep Tapo devices on an isolated IoT/guest network segment separate from corporate assets. Detection is limited since this is a client-side app flaw rather than a network-observable exploit pattern — monitor for unexpected access to device management panels rather than network-based indicators.
What does CISA's SSVC say?
Source: CISA Vulnrichment (SSVC v2.0). Decision based on the CISA Coordinator decision tree.
How is it classified?
Which compliance frameworks are affected?
Compliance analysis pending. Sign in for full compliance mapping when available.
Frequently Asked Questions
What is CVE-2023-27098?
TP-Link's Tapo companion app for smart cameras and plugs ships hardcoded credentials in versions up to 2.12.703, letting anyone who extracts the string from the APK authenticate to the app's login panel without a real password. The flaw is network-exploitable with no privileges or user interaction required (CVSS 7.5), but real-world urgency is limited: EPSS sits at just 0.376% (top 70th percentile, not the high end), it is not in CISA KEV, CISA's SSVC rates it TRACK_STAR (monitor, no immediate action), and no public exploit code or Nuclei scanner template exists. This is a consumer IoT/smart-home app issue, not an AI/ML pipeline vulnerability — it appears in this feed under an 'ml_ui' tag despite having nothing to do with model serving, inference, or AI tooling, likely because Tapo cameras separately market AI-based motion/object detection features unrelated to this specific bug. Organizations that allow TP-Link Tapo devices on corporate or guest networks (e.g., as physical security cameras) should confirm the app is updated past 2.12.703 and keep these devices segmented on an isolated IoT VLAN regardless of patch status.
Is CVE-2023-27098 actively exploited?
No confirmed active exploitation of CVE-2023-27098 has been reported, but organizations should still patch proactively.
How to fix CVE-2023-27098?
Update the Tapo app to the latest release via the official app store — TP-Link's fix removes or rotates the hardcoded credential, though no specific patched version is listed in the CVE record, so confirm against TP-Link's support channel. Until confirmed patched, do not rely on the app's login panel as a security boundary, and keep Tapo devices on an isolated IoT/guest network segment separate from corporate assets. Detection is limited since this is a client-side app flaw rather than a network-observable exploit pattern — monitor for unexpected access to device management panels rather than network-based indicators.
What is the CVSS score for CVE-2023-27098?
CVE-2023-27098 has a CVSS v3.1 base score of 7.5 (HIGH). The EPSS exploitation probability is 0.38%.
What are the technical details?
Original Advisory
TP-Link Tapo APK up to v2.12.703 uses hardcoded credentials for access to the login panel.
Exploitation Scenario
An attacker downloads the public Tapo APK and decompiles it (as demonstrated in the referenced GitHub PoC) to extract the hardcoded credential used by the app's login panel. With network access to a target's Tapo-connected device or companion API — for example on a shared Wi-Fi or guest network — the attacker uses the recovered credential to authenticate without knowing the device owner's real password, gaining access to whatever settings the login panel exposes. No social engineering or user interaction is needed beyond network proximity to the target device.
CVSS Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N References
Timeline
Related Vulnerabilities
CVE-2026-21858 10.0 n8n: Input Validation flaw enables exploitation
Same attack type: Auth Bypass GHSA-vvpj-8cmc-gx39 10.0 picklescan: security flaw enables exploitation
Same attack type: Auth Bypass CVE-2025-2828 10.0 LangChain RequestsToolkit: SSRF exposes cloud metadata
Same attack type: Auth Bypass CVE-2025-53767 10.0 Azure OpenAI: SSRF EoP, no auth required (CVSS 10)
Same attack type: Auth Bypass CVE-2026-26030 10.0 semantic-kernel: Code Injection enables RCE
Same attack type: Auth Bypass