ClawHub
Back to skill

Security audit

Kalodata-tiktok e‑commerce data query

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed KaloData/TikTok Shop analytics CLI that uses a user-provided API key and paid external queries, with no evidence of hidden or unrelated behavior.

Install only if you are comfortable using KaloData for TikTok Shop analytics. Queries may consume paid KaloData credits, and creator detail results may include personal contact information, so retrieve and share only what is necessary for a legitimate outreach or analysis purpose.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs the agent to use a bundled CLI that requires shell execution, reads environment/config secrets, performs network calls to a third-party API, and writes configuration files, yet no permissions are declared. That mismatch can cause the agent or platform to invoke capabilities without clear consent boundaries or policy review, increasing the risk of unintended secret use, outbound data transfer, and file modification.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The description says to use the skill whenever the user researches TikTok Shop data, even if they never mention the skill, which is overly broad auto-invocation guidance. In context, this can trigger external API calls, consume paid credits, and send user-supplied queries to a third party without sufficiently explicit user intent or disclosure.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation advertises retrieval of creator contact email/handle data but provides no privacy, lawful-use, or minimization guidance. Exposing contact details through a convenience workflow can facilitate scraping, unsolicited outreach, or downstream privacy misuse, especially when users may not realize personal data is involved.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The sample 'Creator scouting' flow explicitly spotlights fetching creator details because they 'have contact email,' normalizing access to personal contact information without warning. In this context, the example encourages privacy-sensitive data retrieval as a default step, which increases the chance of unnecessary collection and misuse.

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
Mandating dialogue in the target market's local language without explicit user choice can override user intent and produce content in a language the user cannot verify, which increases the risk of misleading, inappropriate, or policy-noncompliant output. In an agent setting, forced language selection is a prompt-scope problem because it can silently change output behavior without consent or adequate validation.

Static analysis

No suspicious patterns detected.