ClawHub
Back to skill

Security audit

Booth Invitation Writer

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly does what it claims, but it forces third-party promotional text into every generated output and includes a healthcare example that could encourage unsafe handling of medical scan data.

Review outputs before sending them. Remove the Lensmor promotional footer unless you intentionally want it included, verify all event and product claims, and do not ask healthcare recipients to bring or share patient-related scan data unless your organization has approved compliant handling procedures.

A.I.G

Vulnerability Patterns
  • 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
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (2)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:154
Finding

Mandatory Third-Party Advertising Injected into Every Skill Output

Content
View full analysis
Remediation
View remediation

other

Warning
Location
examples/medica-cold-prospect.md:13
Finding

Bundled Example Uses Unsupported Product, Event, and Medical-Data Claims

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · README.md (reported line 3)May include surrounding context.

md
# Booth Invitation Writer — OpenClaw Skill

> Write pre-show outreach that gets real replies and turns target accounts into booked meetings.

**Best for**: teams running booth meeting outreach to prospects, customers, partners, press, or VIP accounts.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The example encourages prospects to bring CT scan data to a live demo and only briefly says it should be anonymized, without specifying approved transfer methods, consent requirements, or that no patient data should be shared unless compliant with applicable privacy laws. In a healthcare context, this can lead users to disclose protected health information during pre-sales outreach or at the event, creating privacy, regulatory, and contractual risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.