ClawHub
Back to skill

Security audit

Trend Demand Forecaster

Security checks for vulnerabilities and agentic risk

Overview

This is a self-contained demand-planning helper with no hidden system access, though some examples should clarify that outside trend data must be user-provided rather than fetched live.

Use this skill as an advisory forecasting template. Provide any trend metrics, marketplace observations, or search-volume exports yourself, and keep purchase orders, budgets, and inventory commitments under human review.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (4)

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

This Chinese scenario directly instructs monitoring multiple live external platforms and using their current signals to predict winning products, which conflicts with the skill's declared offline, heuristic-only scope. In practice, this can push an agent to claim unapproved web access, blend unverifiable live signals into recommendations, and produce risky commercial advice based on fabricated or unsanctioned data gathering.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The title and core description present the skill in Chinese and the example assistant outputs are also in Chinese, while one example user prompt is in English. This suggests the skill may default to Chinese responses without offering the user a language choice or documenting a justified locale constraint.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly states it does not access live systems or external feeds, but this usage scenario suggests deriving forecasts from live Google Trends/search-volume proxies. That contradiction can mislead an agent into fabricating access, overstepping allowed tool boundaries, or presenting externally sourced insights as if they were available within the skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The file is a markdown skill description, so natural-language policy checks apply. Lines L072-L074 introduce instructions and expected output entirely in Chinese, which can impose a specific language/locale experience without any user opt-in or explanation that the skill is region-specific.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.