clawdguard

🦞 ClawdGuard Security hardening for Clawdbot/Moltbot installations. Detects and fixes exposed gateways. Detect and fix exposed Clawdbot/Moltbot gateways Problem • Features • Quick Start • Docker • CLI • What Gets Fixed • Development The Problem 900+ Clawdbot/Moltbot instances are currently expose...

AI agent OpenClaw Arbitrary File Write

The vulnerability in the src/infra/archive.ts script of the AI agent OpenClaw previously known as ClawdBot or MoltBot is related to synchronization errors when using a shared resource. Exploiting this vulnerability allows an attacker to write arbitrary files...

BlueBubbles Plugin Bypass

The vulnerability of the BlueBubbles plugin for the AI agent OpenClaw previously known as ClawdBot or MoltBot is related to deficiencies in the authentication process. Exploiting this vulnerability allows a malicious actor to bypass existing security mechanisms remotely...

From Clawdbot to OpenClaw: Practical Lessons in Building Secure Agents

...

OpenClaw affected by cross-site request forgery (CSRF) through loopback browser mutation endpoints

Summary Browser-facing localhost mutation routes accepted cross-origin browser requests without explicit Origin/Referer validation. Loopback binding reduces remote exposure but does not prevent browser-initiated requests from malicious origins. Impact A malicious website can trigger unauthorized...

GHSA-Q447-RJ3R-2CGH OpenClaw affected by denial of service via unbounded webhook request body buffering

Summary Multiple webhook handlers accepted and buffered request bodies without a strict unified byte/time limit. A remote unauthenticated attacker could send oversized payloads and cause memory pressure, degrading availability. Details Affected packages: - openclaw npm: 2026.2.12 - clawdbot npm:...

OpenClaw affected by denial of service via unbounded webhook request body buffering

Summary Multiple webhook handlers accepted and buffered request bodies without a strict unified byte/time limit. A remote unauthenticated attacker could send oversized payloads and cause memory pressure, degrading availability. Details Affected packages: - openclaw npm: 2026.2.12 - clawdbot npm:...

PT-2026-20368

Name of the Vulnerable Software and Affected Versions OpenClaw versions prior to 2026.2.14 clawdbot versions prior to 2026.1.24-3 Description Browser-facing localhost mutation routes accepted cross-origin browser requests without explicit Origin/Referer validation. Loopback binding reduces remote...

PT-2026-22454

Summary Base64-backed media inputs could be decoded into Buffers before enforcing decoded-size budgets. An attacker supplying oversized base64 payloads can force large allocations, causing memory pressure and denial of service. Attack Scenario Notes - Recommended deployments bind the gateway to...

PT-2026-22451

Summary When multiple Google Chat webhook targets are registered on the same HTTP path, and request verification succeeds for more than one target, inbound webhook events could be routed by first-match semantics. This can cause cross-account policy/context misrouting. Affected Packages / Versions...

PT-2026-22442

Summary Telegram allowlist authorization could match on @username mutable/recyclable instead of immutable numeric sender IDs. Impact Operators who treat Telegram allowlists as strict identity controls could unintentionally grant access if a username changes hands identity rebinding/spoof risk. Th...

PT-2026-22446

Summary Multiple webhook handlers accepted and buffered request bodies without a strict unified byte/time limit. A remote unauthenticated attacker could send oversized payloads and cause memory pressure, degrading availability. Details Affected packages: - openclaw npm: 2026.2.12 - clawdbot npm:...

PT-2026-22439

Summary Archive extraction lacked strict resource budgets, allowing high-expansion ZIP/TAR archives to consume excessive CPU/memory/disk during install/update flows. Affected Packages / Versions - openclaw npm: = 2026.2.13 - clawdbot npm: = 2026.1.24-3 Details Affected component:...

GHSA-3M3Q-X3GJ-F79X OpenClaw optional voice-call plugin: webhook verification may be bypassed behind certain proxy configurations

Affected Packages / Versions This issue affects the optional voice-call plugin only. It is not enabled by default; it only applies to installations where the plugin is installed and enabled. - Package: @openclaw/voice-call - Vulnerable versions: = 2026.2.3 Legacy package name if you are still usi...

PT-2026-22436

Summary Google Chat allowlisting supports matching by sender email in addition to immutable sender resource name users/. This weakens identity binding if a deployment assumes allowlists are strictly keyed by immutable principals. Affected Packages / Versions As of 2026-02-14; based on latest...

A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)

Clawdbot is a self-hosted, tool-using personal AI agent with a broad action space spanning local execution and web-mediated workflows, which raises heightened safety and security concerns under ambiguity and adversarial steering. We present a trajectory-centric evaluation of Clawdbot across six...

OpenClaw < 2026.1.30 Path Traversal (GHSA-r8g4-86fx-92mq)

The version of the OpenClaw AI assistant installed on the remote host is prior to 2026.1.30. It is, therefore, affected by a path traversal vulnerability: - The isValidMedia function in src/media/parse.ts allows arbitrary file paths including absolute paths, home directory paths, and directory...

OpenClaw < 2026.1.20 Command Injection (GHSA-g55j-c2v4-pjcg)

The version of the OpenClaw AI assistant installed on the remote host is prior to 2026.1.20. It is, therefore, affected by a command injection vulnerability: - An unauthenticated local client could use the Gateway WebSocket API to write config via config.apply and set unsafe cliPath values that...

OS Command Injection

clawdbot is vulnerable to an OS command injection. The vulnerability is due to improper escaping and validation of user-supplied input in SSH-related functions, which allows an attacker to inject malicious command strings via the project root path or crafted SSH target arguments, leading to...

Viral AI, Invisible Risks: What OpenClaw Reveals About Agentic Assistants

OpenClaw aka Clawdbot or Moltbot represents a new frontier in agentic AI: powerful, highly autonomous, and surprisingly easy to use. In this research, we examine how its capabilities compare to its predecessors’ and highlight the security risks inherent to the agentic AI paradigm...