Openclaw Bluebubbles 2026.5.7.Tgz

Security checks across malware telemetry and agentic risk

Overview

Prompt-injection indicators were detected in the submitted artifacts (system-prompt-override); human review is required before treating this skill as clean.

Install this only if you want OpenClaw to act as a BlueBubbles/iMessage channel. Before enabling it, protect the BlueBubbles password, restrict allowed senders and group policies, disable unneeded actions, and review catchup/private-network settings. ClawScan detected prompt-injection indicators (system-prompt-override), so this skill requires review even though the model response was benign.

VirusTotal

62/62 vendors flagged this plugin as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
Medium
What this means

If configured and allowed by channel policy, the agent may edit or unsend messages, upload attachments, or change group chat membership through BlueBubbles.

Why it was flagged

The plugin exposes mutating messaging and group-management actions, and they default to enabled when the action config is omitted.

Skill content
edit: z.boolean().default(true), unsend: z.boolean().default(true), ... addParticipant: z.boolean().default(true), removeParticipant: z.boolean().default(true), leaveGroup: z.boolean().default(true), sendAttachment: z.boolean().default(true)
Recommendation

Disable actions you do not need, especially group-management and unsend/edit actions, and restrict allowed senders/groups.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

Enabling this option lets the plugin contact private-network addresses for the BlueBubbles server.

Why it was flagged

The plugin can be configured to reach private/internal network BlueBubbles servers, but the artifact labels this as an explicit dangerous opt-in.

Skill content
Dangerous opt-in for same-host or trusted private/internal BlueBubbles deployments. dangerouslyAllowPrivateNetwork: z.boolean().optional()
Recommendation

Only enable private-network access for a trusted BlueBubbles server URL you control.

#
ASI03: Identity and Privilege Abuse
Medium
What this means

Whoever controls this configuration can let OpenClaw access the BlueBubbles bridge and act on the linked messaging account.

Why it was flagged

The plugin needs a BlueBubbles server password to authenticate to the messaging bridge.

Skill content
serverUrl: z.string().optional(), password: buildSecretInputSchema().optional() ... "password is required when serverUrl is configured"
Recommendation

Store the password as a secret, use the least-exposed BlueBubbles server URL possible, and rotate the password if the configuration is shared.

#
ASI07: Insecure Inter-Agent Communication
Medium
What this means

Incoming message content from BlueBubbles can become agent context and may trigger replies or actions according to your channel policy.

Why it was flagged

Inbound BlueBubbles webhook messages are passed into the agent reply pipeline.

Skill content
BlueBubbles posts JSON to the gateway HTTP server ... Route into core reply pipeline via the plugin runtime
Recommendation

Configure webhook authentication/path carefully and keep allow-from, DM, and group policies limited to trusted senders.

#
ASI08: Cascading Failures
Low
What this means

After a restart or outage, the agent may process recent missed messages rather than only new live messages.

Why it was flagged

Catchup processing is enabled by default and can replay queued BlueBubbles messages through the normal processing pipeline after downtime.

Skill content
Replay messages delivered while the gateway was unreachable. Defaults to on.
Recommendation

Tune or disable catchup if you do not want queued messages processed after downtime.