Back to skill

Security audit

Trend Harvester

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed public trend-research helper, with no executable code or hidden persistence, though its triggers are broad and its language defaults are inconsistent.

Before installing, be aware that generic phrases like 'what's trending' or 'market research' may activate the skill, and the default SKILL.md report template uses Chinese headings. It should only need public web/API access for trend research.

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
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 (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are broad, generic prompts such as 'market research...' and 'what are people saying about...', which can easily appear in ordinary conversation and unintentionally activate the skill. This increases the chance of unexpected tool use, data fetching, or context switching without the user's clear intent, especially in environments where multiple skills compete for activation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The description advertises triggers like 'trending', 'market research', and 'what's hot', which are generic phrases likely to appear in ordinary conversation outside the intended skill context. The file does not provide clear scope limits or exclusion conditions to distinguish when these phrases should or should not invoke the skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The quick usage example is presented as a Chinese command form, and the output template later uses Chinese headings and labels, indicating a default language constraint. The file does not say that users may choose their preferred language or that Chinese output is optional or region-specific.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The mandated output template uses Chinese section titles such as '趋势报告', '核心发现', and '行动建议', which effectively enforces a specific language for responses. No alternative language option or user-choice mechanism is described.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Phrases such as 'what's trending in...', 'market research on...', and especially '热门...' are broad and could match many unrelated user requests. The trigger section does not narrow the context, provide an explicit activation rule, or include negative examples to prevent accidental invocation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.