Back to skill

Security audit

喜悦成长能量

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent parenting-analysis template, but it asks for sensitive child conversations and reproduces them without enough privacy, consent, or non-clinical caution.

Review before installing if you plan to use real child or family conversations. Avoid uploading unnecessary names, school details, health details, or full recordings; prefer short redacted excerpts. Treat the analysis as informal parenting reflection, not clinical assessment, and avoid storing or sharing generated reports that include full transcripts unless all appropriate consent and privacy expectations are clear.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (6)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill is designed to ingest highly sensitive child-parent conversations, recordings, transcripts, and inferred psychological signals, yet it provides no clear warning, consent guidance, or data-handling caution. In this context, omission is dangerous because users may disclose intimate child mental-health and family data without understanding retention, secondary use, or sensitivity risks.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The report template explicitly instructs the system to fully preserve and reproduce the original child-parent conversation, which materially increases exposure of sensitive family content in model outputs. Because the skill handles child data and psychological inference, echoing full transcripts can leak private details into downstream chats, logs, screenshots, exports, or shared reports.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill frames its output as insight into a child's inner world, psychological state, hidden signals, and 'unknown unknowns' without cautioning that such inferences are uncertain and non-clinical. This can lead caregivers to over-trust speculative psychological interpretations, potentially causing harmful parenting decisions or inappropriate labeling of a child.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly solicits audio recordings, transcripts, and detailed child/family conversations, then later states that the full original dialogue will be preserved in the generated report. This creates a clear privacy and data-minimization risk involving highly sensitive child, family, and potentially health-related data, without any warning, consent guidance, or retention limits.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill name, description, headings, examples, and output template are all fixed in Chinese, and no alternative language option or opt-in is described. Under the policy, a skill should not impose a specific language or locale unless it offers a user choice or clearly documents a justified regional constraint.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The display name, description, and the entire skill specification are written only in Chinese, and the output template is also fixed in Chinese. There is no indication that users may choose another language or that the Chinese-only constraint is required for a documented region-specific purpose.

Content

No source excerpt is available for this finding.