ClawHub
Back to skill

Security audit

customer-research

Security checks for vulnerabilities and agentic risk

Overview

This is a customer research guidance skill with no executable code, persistence, or hidden data movement, though users should apply privacy and platform-compliance judgment when mining online comments or handling customer records.

Installers should be comfortable with a skill that helps analyze customer data and public online comments. Before using it on support tickets, interviews, reviews, social profiles, or community posts, minimize personal data, respect platform terms, avoid deanonymizing people, and confirm the target market and locale so conclusions are not overgeneralized.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger description is extremely broad and includes many generic phrases such as 'analyze transcripts,' 'customer sentiment,' and 'what do customers say,' which can cause the skill to activate for unrelated requests. Overbroad activation can route user input and local context into the wrong skill, increasing the risk of inappropriate file reads, misleading outputs, or unintended handling of sensitive business data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The guide gives detailed instructions for mining user-generated content, reviews, comments, and audience intelligence across many third-party platforms, but it provides no privacy, consent, retention, or terms-of-service safeguards. In a customer-research skill, that omission can normalize indiscriminate collection, copying, and synthesis of personal or quasi-personal data, increasing legal, policy, and reputational risk when operators aggregate quotes or profiles at scale.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

This natural-language statement acknowledges an English/US-centric constraint in the research approach, which can amount to a locale bias if used as default guidance without opt-in or mitigation. The file does not pair this limitation with instructions to offer language/region choice or validate findings in other locales.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.