Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessMove within BLE range of target robot
-
DeliverySend crafted BLE GATT writes to SSID chunk accumulator
-
ExploitOverflow fixed BSS buffer with attacker-controlled payload
-
ExecutionCorrupt adjacent mainloop function pointer dispatch entry
-
PersistTrigger BLE connection cleanup path
-
ImpactInvoke corrupted pointer, passing attacker data to system() as uid 0
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V15 - Secure Coding and Architecture Bounds checking, write limits
MITRE ATT&CK
2 techniques across 2 tactics: Execution, Privilege Escalation The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Attackers must be within BLE proximity of the target robot, typically under 10 meters and line-of-sight dependent. No pairing, authentication, or credentials of any kind are required - the BLE GATT server accepts writes from unpaired, unauthenticated devices, which is the condition enabling PR:N. The robot must be running firmware through version 1.5.2 with the BLE GATT server active and the WiFi provisioning stack loaded; available data indicates this is the default operational configuration, not a non-default mode. Exploitation requires successfully executing a multi-step chain - overflowing the BSS buffer, achieving precise corruption of the adjacent function pointer, and triggering the cleanup path - which is why AC:H is assigned; single-packet exploitation is not indicated by the description. No user interaction on the robot side is required. Limiting factors include the physical proximity constraint, the high chaining complexity, and the requirement that BLE remains active and the provisioning stack is running at time of attack. |
| Impact | Real-world risk is moderate-to-significant for any environment deploying Unitree G1 EDU robots within BLE range of untrusted parties. The CVSS 4.0 score of 7.7 (AV:A/AC:H/PR:N/UI:N) correctly reflects that exploitation requires physical proximity - BLE range, typically under 10 meters - and involves a high-complexity multi-step chain, which meaningfully limits opportunistic mass exploitation. However, the publicly available UniBLEed proof-of-concept on GitHub dramatically lowers the skill threshold, and the end result is an unconditional root shell on the robot platform. The vulnerability is not in CISA KEV and no EPSS data was provided, suggesting exploitation has not yet been observed at scale; nonetheless, the gap between KEV absence and public POC availability is a meaningful warning sign that targeted exploitation may emerge before widespread detection. Research institutions, warehouses, and manufacturing environments where G1 EDU units operate near untrusted personnel should prioritize physical access controls and firmware updates, given the no-authentication, no-pairing exploitation path. |
| Exploit ScenarioDerived | An attacker positions themselves within BLE range of an operational Unitree G1 EDU robot - no pairing, account, or prior access required - and uses the UniBLEed tool to send a sequence of crafted BLE GATT writes targeting the SSID chunk accumulator, overflowing the fixed BSS buffer and overwriting the adjacent function pointer dispatch table. When the BLE connection cleanup path fires, the robot firmware executes attacker-supplied shell commands as root, yielding full persistent control of the robotic platform. Because the proof-of-concept is publicly available, this attack is within reach of non-expert adversaries who can achieve physical proximity. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 27, 2026
Root code execution in Unitree G1 EDU robot firmware through version 1.5.2 is achievable by proximate unauthenticated attackers via a chained BLE GATT and WiFi provisioning exploit requiring no pairing or credentials. Crafted BLE writes overflow a fixed BSS buffer in the SSID chunk accumulator, corrupting an adjacent mainloop function pointer dispatch entry; the cleanup path subsequently invokes this corrupted pointer, passing attacker-controlled data to system() as uid 0. A public proof-of-concept (UniBLEed) is available, and the CVSS 4.0 score of 7.7 reflects the adjacent attack vector and high complexity of the exploit chain.
RemediationAI
No vendor-released patch identified at time of analysis - no fixed firmware version is specified in any available reference. As compensating controls, operators should disable or restrict BLE functionality on G1 EDU units where BLE provisioning is not operationally required; note that disabling BLE may impair the robot commissioning workflow and should be evaluated against operational needs. Physical access controls limiting untrusted personnel to beyond BLE range (approximately 10 meters) directly reduce exposure for the adjacent-vector attack. Network segmentation of the WiFi provisioning environment can limit the impact of the heredoc injection component of the chain. Monitor for firmware updates from Unitree Robotics and track the VulnCheck advisory at https://www.vulncheck.com/advisories/unitree-g1-edu-ble-gatt-rce-via-wifi-provisioning-stack. The public exploit repository at https://github.com/OlivierLaflamme/UniBLEed can serve as a regression test to validate that any future firmware release resolves the vulnerability before deployment.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The vulnerability chain targets two weaknesses in Unitree G1 EDU firmware (CPE: cpe:2.3:a:unitree_robotics:g1_edu:*:*:*:*:*:*:*:*): an unquoted heredoc variable in the WiFi provisioning shell script enabling command injection, and a CWE-787 out-of-bounds write in the BLE GATT server's SSID chunk accumulator. The accumulator writes attacker-controlled BLE packet payloads into a statically allocated BSS buffer without adequate bounds checking. Because BSS memory layout places a mainloop function pointer dispatch table adjacent to this buffer, overflow corrupts that dispatch entry. The cleanup code path subsequently calls through the corrupted pointer, with attacker-supplied data passed directly to system() running as uid 0. This chain is enabled by the absence of bounds validation on BLE-received SSID data and insufficient hardening (e.g., no stack canaries or ASLR enforcement) in the firmware build.
Know when G1 Edu gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:A/AC:H/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
Adjacent vector for BLE proximity requirement; AC:H for multi-step chained exploit; PR:N as no pairing or credentials required; full C/I/A from unconditional root shell; scope unchanged as impact is contained to the robot.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
DescriptionCVE.org
Unitree G1 EDU firmware through 1.5.2 contains multiple chained vulnerabilities in the BLE GATT server and WiFi provisioning stack that allow unauthenticated proximate attackers to achieve root code execution without pairing or credentials by exploiting an unquoted heredoc variable in the WiFi provisioning script and a buffer overflow in the SSID chunk accumulator. Attackers can send crafted BLE writes to overflow a fixed BSS buffer across BLE connections, corrupting an adjacent mainloop function pointer dispatch entry that is subsequently invoked by the cleanup path passing attacker-controlled data to system() as uid 0.
Latest POC
1 public PoC link from github
Sign in to open the public PoC links collected for this CVE.
Executive ViewAI
Unitree G1 EDU robots running firmware through version 1.5.2 are vulnerable to unauthenticated root code execution via chained Bluetooth and WiFi exploits, with publicly available exploit code existing. Although attacks require physical proximity, the absence of any vendor patch means all affected robots should be considered operationally at-risk and require immediate isolation, network restriction, or decommissioning until updates are released.
No vendor-released patch identified at time of analysis. Interim mitigations: (1) Permanently disable WiFi provisioning and restrict network access to hardened, isolated management VLANs with strict egress filtering; (2) Deploy Bluetooth signal shielding or restrict affected robots to physically secured, access-controlled areas to eliminate proximity attack surface; (3) Implement network segmentation and application-level access controls preventing compromised robots from reaching production data or systems; (4) Establish continuous behavioral monitoring and real-time alerting for anomalous communications or state changes; (5) Maintain offline configuration backups and implement manual approval processes for all critical robot operations.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
A buffer overflow occurs when a program writes more data to a memory buffer than it was allocated to hold, causing the excess data to spill into adjacent memory regions.
Memory corruption occurs when a program writes data beyond the boundaries of allocated memory regions or accesses memory in unintended ways, violating the integrity of the process's address space.
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
EUVD-2026-67166
GHSA-4qj2-mfrv-5fm4
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB