ClawHub
Back to skill

Security audit

Voice of Customer

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Voice of Customer analysis guide, but it encourages aggregating support tickets, chat transcripts, surveys, reviews, and social mentions without privacy, redaction, or access-control guidance.

Only use this skill with approved feedback sources and sanitized datasets. Remove or anonymize names, emails, phone numbers, account IDs, order IDs, payment details, health information, and other sensitive content before analysis, and avoid placing raw support transcripts or verbatim customer quotes into unsecured tools or broad reports.

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

SkillSpector

Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · SKILL.md (reported line 129)May include surrounding context.

md
| Shipping damage | 12% | 2.1/5 | ↑ Increasing | "Bottle arrived cracked and leaked all over the box" |
| Results exceeded expectations | 11% | 4.8/5 | Stable | "After 3 weeks my dark spots have noticeably faded" |
| Subscription flexibility | 9% | 2.4/5 | Stable | "I can't skip a month without calling support" |
| Price increase frustration | 8% | 1.9/5 | ↑ New | "Went up $8 with no warning or explanation" |

**Step 5 — Priority Matrix:**

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill explicitly instructs users to aggregate support tickets, chat transcripts, surveys, reviews, and social media mentions into a unified dataset, but it provides no guidance on handling personal data, sensitive content, access controls, minimization, or redaction. In practice, those sources often contain PII, account details, health information, or other confidential customer data, so omitting privacy and data-handling safeguards can lead to unauthorized exposure or unsafe downstream processing.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.