Nanobot CVE-2026-35589

nanobot versions prior to 0.1.5 with WebSocket bridge server running on localhost (127.0.0.1). BRIDGE_TOKEN authentication must be disabled (default). Victim must click attacker-controlled link while logged into nanobot. Requires browser-based attack; origin header validation absent.

Real-world risk is moderate despite the 8.0 CVSS score. The attack requires high complexity (AC:H) and user interaction (UI:R), meaning an attacker must craft a malicious website and social-engineer the victim into visiting it while their nanobot bridge is actively running. The scope change (S:C) and high confidentiality/integrity impacts (C:H/I:H) reflect the serious consequences if successfully exploited: complete WhatsApp session control, message interception, and credential theft. However, several factors reduce practical exploitability: the victim must be running the nanobot bridge locally, the attacker must time the attack to coincide with active bridge usage, and modern browser security features may provide some defense-in-depth. No EPSS score or KEV status is provided, and no public exploit identified at time of analysis suggests limited active exploitation. The fact this is an incomplete fix of a prior CVE indicates the vendor is responsive to security issues. SSVC framework considerations: exploitation status appears proof-of-concept at most, technical impact is total (session hijacking), but automatable exploitation is limited by the need for precise timing and user interaction. Priority should be high for organizations using nanobot in production, but this is not a wormable or mass-exploitation threat.

An attacker creates a malicious website containing JavaScript that attempts to connect to ws://127.0.0.1:3001/ when loaded. The victim, who has nanobot bridge running locally for WhatsApp integration, visits this site through social engineering (phishing email, compromised legitimate site, or malvertising). Because the bridge server lacks Origin validation, the cross-origin WebSocket connection succeeds, granting the attacker's JavaScript full access to the bridge API. The attacker can now read incoming WhatsApp messages in real-time, steal QR code authentication data to clone the session, and send messages impersonating the victim.