Security audit

LobsterHub Bridge

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent setup guide, but it asks users to connect their local AI gateway to an external service with weak disclosure and insecure transport.

Review carefully before installing. Use this only if you are comfortable allowing a third-party plugin and service to relay other users' messages to your local OpenClaw AI. Prefer a pinned and reviewed plugin version, HTTPS/WSS endpoints, localhost-only gateway binding, a dedicated no-tools assistant profile, strict rate limits, and revocable short-lived pairing credentials.

A.I.G

Vulnerability Patterns

Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions

Findings (3)

T08 · Insecure Dependencies

Error

Location: SKILL.md:30
Finding: Unpinned and Unreviewable Third-Party Plugin Installation
Content: View full analysis
Remediation: View remediation

T05 · Unauthorized Access and Privilege Escalation

Error

Location: SKILL.md:18
Finding: Externally Controlled Messages Are Bridged to a Local AI Gateway Without Documented Isolation
Content: View full analysis
Remediation: View remediation

T09 · Insecure Skill Coding Practices

Error

Location: SKILL.md:46
Finding: Plaintext Remote Service Endpoints Expose Account and Bridge Communications
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Missing User Warnings

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill instructs users to enable the Gateway HTTP API, install a third-party plugin, and connect to an external host and WebSocket bridge, but it does not clearly warn that this exposes the local AI service to external registration and message relay traffic. The repeated privacy reassurance ('your data stays private') combined with plain HTTP links to an IP address makes the omission more dangerous because it may lower user suspicion while encouraging trust in a remote service.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 85% confidence
Finding: The skill metadata advertises broad, user-invocable functionality without clear trigger boundaries or safety constraints, which can cause users to invoke it casually without understanding that it leads to installation of a network-connected plugin and exposure of local services. In this context, the vague description increases the chance of unsafe use because the skill ultimately facilitates external connectivity to a local AI gateway.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.