OpenClaw is a personal AI assistant. OpenClaw macOS desktop client registers the `openclaw://` URL scheme. For `openclaw://agent` deep links without an unattended `key`, the app shows a confirmation dialog that previously displayed only the first 240 characters of the message, but executed the full message after the user clicked "Run." At the time of writing, the OpenClaw macOS desktop client is still in beta. In versions 2026.2.6 through 2026.2.13, an attacker could pad the message with whitespace to push a malicious payload outside the visible preview, increasing the chance a user approves a different message than the one that is actually executed. If a user runs the deep link, the agent may perform actions that can lead to arbitrary command execution depending on the user's configured tool approvals/allowlists. This is a social-engineering mediated vulnerability: the confirmation prompt could be made to misrepresent the executed message. The issue is fixed in 2026.2.14. Other mitigations include not approve unexpected "Run OpenClaw agent?" prompts triggered while browsing untrusted sites and usingunattended deep links only with a valid `key` for trusted personal automations.

OpenClaw versions 2026.1.29-beta.1 prior to 2026.2.1 contain a path traversal vulnerability in plugin installation that allows malicious plugin package names to escape the extensions directory. Attackers can craft scoped package names containing path traversal sequences like .. to write files outside the intended installation directory when victims run the plugins install command.

OpenClaw before 2026.5.6 contains an authorization bypass vulnerability in Telegram interactive callbacks that allows authenticated users to skip commands.allowFrom validation. Attackers can invoke affected callbacks to mark themselves as authorized senders before allowlist checks are applied, triggering command behavior outside configured Telegram sender restrictions.

OpenClaw before 2026.4.25 contains a policy bypass vulnerability in embedded runner policy that allows requests using provider aliases to compare against aliases instead of canonical provider identities. Attackers can exploit this confusion to select bundled tool access outside intended provider policy restrictions when the affected feature is enabled.

OpenClaw before 2026.4.20 contains a tool policy bypass vulnerability allowing bundled MCP and LSP tools to circumvent configured tool restrictions. Attackers with local agent access can append restricted tools to the effective tool set after policy filtering, bypassing profile policies, allow/deny lists, owner-only restrictions, sandbox policies, and subagent policies.

OpenClaw before 2026.4.21 contains an authorization bypass vulnerability in command-auth.ts that allows non-owner senders to execute owner-enforced slash commands when wildcard inbound senders are configured without explicit owner allowFrom settings. Attackers can exploit this by sending commands like /send, /config, or /debug on affected channels to bypass owner-only command authorization checks.

OpenClaw before 2026.4.20 contains a hook session-key bypass vulnerability that allows attackers to circumvent the hooks.allowRequestSessionKey opt-in restriction. Attackers can render externally influenced session keys through templated hook mappings to bypass webhook routing isolation controls.

OpenClaw before 2026.4.8 contains a role bypass vulnerability in the device.token.rotate function that allows minting tokens for unapproved roles. Attackers can bypass device role-upgrade pairing to preserve or mint roles and scopes that had not undergone intended approval.

OpenClaw before 2026.4.20 contains an improper authorization vulnerability in paired-device pairing management that allows limited-scope sessions to enumerate and act on pairing requests. Attackers with paired-device access can approve or operate on unrelated pending device requests within the same gateway scope.

OpenClaw before 2026.4.8 contains an improper authorization vulnerability where the node.pair.approve method accepts operator.write scope instead of the narrower operator.pairing scope, allowing unprivileged users to approve node pairing. Attackers with operator.write permissions can bypass pairing approval restrictions to gain unauthorized access to exec-capable nodes.

OpenClaw before 2026.4.8 contains a privilege escalation vulnerability in the gateway plugin HTTP authentication mechanism that escalates identity-bearing operator.read requests to runtime operator.write permissions. Attackers can exploit this by sending read-scoped requests through the gateway auth route to gain unauthorized write access to runtime operations.

OpenClaw before 2026.4.8 contains a security bypass vulnerability in node.invoke(browser.proxy) that allows mutation of persistent browser profiles. Attackers can exploit this path to circumvent the browser.request persistent profile-mutation guard and modify browser configurations.

OpenClaw before 2026.4.8 contains a privilege escalation vulnerability allowing previously paired nodes to reconnect with exec-capable commands without the operator.admin scope requirement. Attackers can bypass re-pairing authentication to execute privileged commands on the local assistant system.

OpenClaw versions 2026.4.5 before 2026.4.10 contain a sandbox escape vulnerability allowing sandboxed agents to override exec routing by specifying host=node. Attackers can bypass sandbox boundaries and route execution to remote nodes instead of intended sandbox paths.

OpenClaw versions 2026.2.23 before 2026.4.12 contain a weakened exec approval binding vulnerability in busybox and toybox applet execution that allows attackers to obscure which applet would actually run. Attackers can exploit opaque multi-call binaries to bypass exec approval mechanisms and weaken risk classification of unsafe applet invocations.

OpenClaw before 2026.3.28 contains a privilege escalation vulnerability in the chat.send endpoint that allows write-scoped gateway callers to persist admin-only verboseLevel session overrides. Attackers can exploit the /verbose parameter to bypass access controls and expose sensitive reasoning or tool output intended to be restricted to administrators.

OpenClaw before 2026.3.31 contains a session visibility bypass vulnerability where the session_status function fails to enforce configured tools.sessions.visibility restrictions for unsandboxed invocations. Attackers can invoke session_status without sandbox constraints to bypass session-policy controls and access restricted session information.

OpenClaw before 2026.3.28 contains a privilege escalation vulnerability in chat.send that allows write-scoped gateway callers to trigger admin-only session reset operations. Attackers can rotate target sessions, archive prior transcript state, and force new session IDs without requiring admin scope by exploiting improper authorization checks in the chat.send path.

OpenClaw before 2026.3.31 contains an access control bypass vulnerability in the Discord voice manager that allows attackers to bypass channel-level member access allowlist restrictions. Attackers can send Discord voice ingress requests before channel allowlist authorization is performed, gaining unauthorized access to restricted voice channels.

OpenClaw before 2026.3.25 contains an authorization bypass vulnerability in the HTTP /sessions/:sessionKey/history route that skips operator.read scope validation. Attackers can access session history without proper operator read permissions by sending HTTP requests to the vulnerable endpoint.

OpenClaw versions prior to 2026.2.26 fail to enforce sender authorization in member and message subtype system event handlers, allowing unauthorized events to be enqueued. Attackers can bypass Slack DM allowlists and per-channel user allowlists by sending system events from non-allowlisted senders through message_changed, message_deleted, and thread_broadcast events.

OpenClaw before 2026.4.15 contains an authorization bypass vulnerability in Matrix room control-command authorization that trusts DM pairing-store entries. Attackers with DM-paired sender IDs can execute room control commands without being in configured allowlists by posting in bot rooms, potentially enabling privileged OpenClaw behavior.

OpenClaw before 2026.3.12 contains an insufficient access control vulnerability in the /config and /debug command handlers that allows command-authorized non-owners to access owner-only surfaces. Attackers with command authorization can read or modify privileged configuration settings restricted to owners by exploiting missing owner-level permission checks.

OpenClaw before 2026.3.11 contains an authorization bypass vulnerability allowing write-scoped callers to reach admin-only session reset logic. Attackers with operator.write scope can issue agent requests containing /new or /reset slash commands to reset targeted conversation state without holding operator.admin privileges.

OpenClaw before 2026.3.11 contains an authorization bypass vulnerability allowing authenticated operators with only operator.write permission to access admin-only browser profile management routes through browser.request. Attackers can create or modify browser profiles and persist attacker-controlled remote CDP endpoints to disk without holding operator.admin privileges.

OpenClaw before 2026.3.11 contains an approval integrity vulnerability where system.run approvals fail to bind mutable file operands for certain script runners like tsx and jiti. Attackers can obtain approval for benign script commands, rewrite referenced scripts on disk, and execute modified code under the approved run context.

OpenClaw before 2026.3.28 contains a privilege escalation vulnerability in the /pair approve command path that fails to forward caller scopes into the core approval check. A caller with pairing privileges but without admin privileges can approve pending device requests asking for broader scopes including admin access by exploiting the missing scope validation in extensions/device-pair/index.ts and src/infra/device-pairing.ts.

OpenClaw versions prior to 2026.2.22 contain an authentication bypass vulnerability that allows clients authenticated with a shared gateway token to connect as role=node without device identity verification. Attackers can exploit this by claiming the node role during WebSocket handshake to inject unauthorized node.event calls, triggering agent.request and voice.transcript flows without proper device pairing.

OpenClaw versions prior to 2026.2.25 fail to enforce sender authorization checks for interactive callbacks including block_action, view_submission, and view_closed in shared workspace deployments. Unauthorized workspace members can bypass allowFrom restrictions and channel user allowlists to enqueue system-event text into active sessions.

OpenClaw versions prior to 2026.2.26 contain an authorization bypass vulnerability where DM pairing-store identities are incorrectly treated as group allowlist identities when dmPolicy=pairing and groupPolicy=allowlist. Remote attackers can send messages and reactions as DM-paired identities without explicit groupAllowFrom membership to bypass group sender authorization checks.

OpenClaw versions prior to 2026.2.26 contain an authorization bypass vulnerability where DM pairing-store identities are incorrectly eligible for group allowlist authorization checks. Attackers can exploit this cross-context authorization flaw by using a sender approved via DM pairing to satisfy group sender allowlist checks without explicit presence in groupAllowFrom, bypassing group message access controls.

OpenClaw versions prior to 2026.2.25 fail to enforce dmPolicy and allowFrom authorization checks on Discord direct-message reaction notifications, allowing non-allowlisted users to enqueue reaction-derived system events. Attackers can exploit this inconsistency by reacting to bot-authored DM messages to bypass DM authorization restrictions and trigger downstream automation or tool policies.

OpenClaw versions prior to 2026.3.2 fail to pass the senderIsOwner flag when processing Discord voice transcripts in agentCommand, causing the flag to default to true. Non-owner voice participants can exploit this omission to access owner-only tools including gateway and cron functionality in mixed-trust channels.

OpenClaw versions 2026.2.22 prior to 2026.2.25 contain a privilege escalation vulnerability allowing unpaired device identities to bypass operator pairing requirements and self-assign elevated operator scopes including operator.admin. Attackers with valid shared gateway authentication can present a self-signed unpaired device identity to request and obtain higher operator scopes before pairing approval is granted.

OpenClaw versions prior to 2026.3.1 contain an authorization mismatch vulnerability that allows authenticated callers with operator.write scope to invoke owner-only tool surfaces including gateway and cron through agent runs in scoped-token deployments. Attackers with write-scope access can perform control-plane actions beyond their intended authorization level by exploiting inconsistent owner-only gating during agent execution.

OpenClaw versions prior to 2026.2.25 contain an access control vulnerability in signal reaction notification handling that allows unauthorized senders to enqueue status events before authorization checks are applied. Attackers can exploit the reaction-only event path in event-handler.ts to queue signal reaction status lines for sessions without proper DM or group access validation.

OpenClaw versions prior to 2026.2.26 contain an authorization bypass vulnerability in the pairing-store access control for direct message pairing policy that allows attackers to reuse pairing approvals across multiple accounts. An attacker approved as a sender in one account can be automatically accepted in another account in multi-account deployments without explicit approval, bypassing authorization boundaries.

OpenClaw version 2026.2.22-2 prior to 2026.2.23 tools.exec.safeBins validation for sort command fails to properly validate GNU long-option abbreviations, allowing attackers to bypass denied-flag checks via abbreviated options. Remote attackers can execute sort commands with abbreviated long options to skip approval requirements in allowlist mode.

OpenClaw versions prior to 2026.2.14 contain a vulnerability in the gateway in which it fails to sanitize internal approval fields in node.invoke parameters, allowing authenticated clients to bypass exec approval gating for system.run commands. Attackers with valid gateway credentials can inject approval control fields to execute arbitrary commands on connected node hosts, potentially compromising developer workstations and CI runners.

OpenClaw versions prior to 2026.2.2 contain an authorization bypass vulnerability where clients with operator.write scope can approve or deny exec approval requests by sending the /approve chat command. The /approve command path invokes exec.approval.resolve through an internal privileged gateway client, bypassing the operator.approvals permission check that protects direct RPC calls.

OpenClaw's Nextcloud Talk plugin versions prior to 2026.2.6 accept equality matching on the mutable actor.name display name field for allowlist validation, allowing attackers to bypass DM and room allowlists. An attacker can change their Nextcloud display name to match an allowlisted user ID and gain unauthorized access to restricted conversations.

OpenClaw versions prior to 2026.3.7 contain a shell approval gating bypass vulnerability in system.run dispatch-wrapper handling that allows attackers to skip shell wrapper approval requirements. The approval classifier and execution planner apply different depth-boundary rules, permitting exactly four transparent dispatch wrappers like repeated env invocations before /bin/sh -c to bypass security=allowlist approval gating by misaligning classification with execution planning.

OpenClaw versions prior to 2026.3.7 contain a sandbox escape vulnerability in the /acp spawn command that allows authorized sandboxed sessions to initialize host-side ACP runtime. Attackers can bypass sandbox restrictions by invoking the /acp spawn slash-command to cross from sandboxed chat context into host-side ACP session initialization when ACP is enabled.

OpenClaw is a personal AI assistant. Prior to 2026.2.13, the optional BlueBubbles iMessage channel plugin could accept webhook requests as authenticated based only on the TCP peer address being loopback (`127.0.0.1`, `::1`, `::ffff:127.0.0.1`) even when the configured webhook secret was missing or incorrect. This does not affect the default iMessage integration unless BlueBubbles is installed and enabled. Version 2026.2.13 contains a patch. Other mitigations include setting a non-empty BlueBubbles webhook password and avoiding deployments where a public-facing reverse proxy forwards to a loopback-bound Gateway without strong upstream authentication.

OpenClaw versions prior to 2026.2.22 with the optional BlueBubbles plugin contain an access control bypass vulnerability where empty allowFrom configuration causes dmPolicy pairing and allowlist restrictions to be ineffective. Remote attackers can send direct messages to BlueBubbles accounts by exploiting the misconfigured allowlist validation logic to bypass intended sender authorization checks.

OpenClaw before 2026.4.8 omits owner-only enforcement for cross-channel allowlist writes in the /allowlist endpoint. An authorized non-owner sender can bypass access controls to perform allowlist modifications against different channels, violating the intended trust model.

OpenClaw before 2026.4.20 contains a scope enforcement bypass vulnerability in the assistant-media route that allows trusted-proxy callers without operator.read scope to access protected assistant-media files and metadata. Attackers can bypass identity-bearing HTTP auth path scope validation to retrieve sensitive media content within allowed media roots.

OpenClaw versions 2026.4.9 before 2026.4.10 contain a sender policy bypass vulnerability in the outbound host-media attachment read helper that allows unauthorized local file disclosure. Attackers with denied read access via toolsBySender or group policy can trigger host-media attachment loading to bypass sender and group-scoped authorization boundaries and retrieve readable local files through the outbound media path.

OpenClaw before 2026.3.31 contains an authorization bypass vulnerability in Discord slash command and autocomplete paths that fail to enforce group DM channel allowlist restrictions. Authorized Discord users can bypass channel restrictions by invoking slash commands, allowing access to restricted group DM channels.

OpenClaw versions 2026.2.14 through 2026.3.24 fail to consistently apply guild and channel policy gates to Discord button and component interactions. Attackers can trigger privileged component actions from blocked contexts by bypassing channel policy enforcement.