Back to skill

Security audit

Growth Guardian

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent parenting journal, but it stores and analyzes highly sensitive information about children without enough privacy safeguards.

Review before installing. Use aliases instead of real child names, avoid storing raw recordings or full transcripts unless necessary, choose a private folder, and periodically delete records you no longer need. Treat the psychological observations as non-diagnostic and confirm before any file is created or updated.

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
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 4)May include surrounding context.

md
---
name: growth-guardian
displayName: 成长守护
description: >-
  儿童成长守护 Skill。基于8维发展模型(D0-D7)结构化记录成长档案,
  从对话中洞察心理状态,生成周报/月报/季度评估。
  触发词:记录孩子/成长记录/发展评估/周报/孩子最近怎么样/对话分析/育儿观察/内驱力
version: "2.0.0"
type: skill
tags: ["育儿", "儿童成长", "8维模型", "对话分析", "内在动机", "成长守护"]
author: duzhilei951
---

# 🌱 成长守护

## 从记录到洞察,守护孩子成长的每一步 💎

---

## 是什么

「成长守护」是一个专为家长设计的 AI 成�

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill instructs users to provide and store child identifiers, age, family situation, and ongoing developmental observations, but does not present a clear privacy notice, consent boundary, retention policy, or warning about the sensitivity of minors' data. Because the subject is a child, the context materially increases the severity of collecting persistent personal and behavioral records without explicit safeguards.

Content

No source excerpt is available for this finding.

Ssd 3

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill explicitly directs the agent to gather and persist sensitive child and family information, including name, age, household members, personality observations, interests, and ongoing developmental records, into structured files. Persisting this volume of minors' personal data increases the risk of privacy harm, unauthorized exposure, profiling, and long-term misuse, especially when no minimization or protection controls are specified.

Content

No source excerpt is available for this finding.

Ssd 3

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill directs the system to ingest child conversation transcripts or audio-derived text and infer 'hidden' psychological signals for ongoing analysis and model mapping. This creates a sensitive profiling pipeline around a minor, with risk of overcollection, inaccurate psychological labeling, and harmful persistence of intimate behavioral inferences.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill invites users to submit audio files, transcripts, and direct child dialogue for psychological interpretation, including inference about inner state and hidden signals, without a strong warning about privacy, consent, or the limits of such inference. This is especially sensitive because it processes minors' conversations and derives subjective psychological conclusions that could be stored and reused over time.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases include very common parenting-language queries such as '孩子最近怎么样' and '周报', which can cause the skill to activate in ordinary conversations without clear user intent. In this skill, unintended invocation is more dangerous because activation leads into workflows that collect and analyze sensitive child, family, and psychological data.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The display name, description, prompts, examples, directory names, and output formats are entirely specified in Chinese, with no indication that users may choose another language. This can violate language/locale policy when a skill imposes a single language without explicit opt-in or documented regional scope.

Content

No source excerpt is available for this finding.