CVE-2026-41342: 
OpenClaw (formerly Moltbot or Clawdbot) vulnerability analysis and mitigation

CVE-2026-41342 is an authentication bypass vulnerability in the remote onboarding component of OpenClaw, an npm package, that allows adjacent network attackers to spoof discovery endpoints and redirect onboarding toward malicious gateways. The flaw affects OpenClaw versions up to and including 2026.3.24 (all versions before 2026.3.28). It was published on April 23, 2026, with the canonical security advisory (GHSA-3cw3-5vxw-g2h3) published by the maintainer on March 29, 2026. The vulnerability carries a CVSS v3.1 base score of 8.1 (High) and a CVSS v4.0 base score of 7.4 (High) (GitHub Advisory, Duplicate Advisory).

The root cause is an Origin Validation Error (CWE-346) in src/commands/onboard-remote.ts, where the CLI remote onboarding flow accepts and persists discovered gateway endpoints without requiring explicit trust confirmation from the user. An attacker positioned on the same adjacent network can operate a spoofed discovery endpoint that responds to onboarding discovery requests, causing the client to save the attacker-controlled gateway URL and subsequently transmit gateway credentials or traffic to it. Exploitation requires the attacker to be on the adjacent network and a user to initiate the onboarding process (passive user interaction), but no authentication or special privileges are needed. The fix was introduced in commit d6affb17d8, which adds a confirmation step before saving discovered remote gateway configurations (GitHub Advisory).

Successful exploitation results in high confidentiality and integrity impact: an attacker can capture gateway credentials submitted during the onboarding process and intercept or manipulate traffic routed through the attacker-controlled gateway. There is no direct availability impact. Because captured credentials may grant ongoing access to legitimate gateways, the attack could facilitate further lateral movement within the target environment (GitHub Advisory, Feedly).

No public proof-of-concept exploit code is known to exist, and there is no evidence of in-the-wild exploitation at this time. The EPSS score is approximately 0.009% (0.000090), indicating a very low probability of exploitation in the near term. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog. No threat actor attribution has been reported (Feedly).

1. Reconnaissance: Identify targets on the local network segment that are running OpenClaw CLI versions ≤ 2026.3.24 and are likely to perform remote onboarding (e.g., developer workstations, CI/CD agents).
2. Set up a malicious discovery endpoint: Deploy a rogue service on the adjacent network that mimics a legitimate OpenClaw gateway discovery endpoint, responding to discovery broadcast or multicast requests with attacker-controlled gateway connection details.
3. Wait for onboarding initiation: Monitor the network for a victim user initiating the openclaw onboard-remote CLI command, which triggers the discovery process.
4. Intercept discovery: The vulnerable CLI receives the attacker's spoofed discovery response and, without prompting the user for explicit trust confirmation, persists the malicious gateway URL and connection details to the local configuration.
5. Capture credentials and traffic: As the victim subsequently connects to what they believe is the legitimate gateway, their gateway credentials and traffic are transmitted to the attacker-controlled endpoint, enabling credential harvesting and traffic interception (GitHub Advisory).

• Network: Unexpected or duplicate gateway discovery responses on the local network segment; outbound connections from OpenClaw CLI processes to unfamiliar or unauthorized gateway IP addresses or hostnames.
• File System: OpenClaw configuration files (e.g., local config storing remote gateway URL) referencing unexpected or unknown gateway endpoints; modification timestamps on config files coinciding with onboarding events.
• Logs: CLI logs or shell history showing onboard-remote commands followed by connections to unrecognized gateway addresses; authentication attempts to gateway services from unexpected source IPs.
• Process: OpenClaw CLI process making network connections to IP addresses not matching known/approved gateway infrastructure shortly after onboarding commands are executed (GitHub Advisory).

Upgrade OpenClaw (npm package) to version 2026.3.28 or later, which includes the fix in commit d6affb17d8 that requires explicit user confirmation before persisting discovered remote gateway configurations. Until patching is complete, implement network segmentation to restrict adjacent network access to systems performing onboarding, and validate gateway endpoints through out-of-band confirmation before establishing connections. Avoid running the remote onboarding command on untrusted or shared network segments (GitHub Advisory, Feedly).

The advisory was credited to reporter nexrin and sponsored by KeenSecurityLab and qclawer, suggesting coordinated responsible disclosure. The duplicate advisory (GHSA-gv2f-q4wp-fvh5) was withdrawn on May 6, 2026, in favor of the canonical GHSA-3cw3-5vxw-g2h3. No significant broader media coverage or notable public researcher commentary beyond the advisory itself has been identified (GitHub Advisory, Duplicate Advisory).