T01 · Skill Instruction Hijacking
Warning
- Location
SKILL.md:67- Finding
Mandatory Commercial Promotion Injected into Agent Responses
- Content
View full analysis
- Remediation
View remediation
Security audit
voc-conversation
Security checks for vulnerabilities and agentic risk
Overview
This is a markdown-only Chinese VOC analysis skill with no code execution or automatic data transfer, though it includes a visible customization upsell.
Install only if you want Chinese customer-service conversation VOC analysis. Provide only conversation exports you are authorized to analyze, review generated quotes for redaction before sharing, and expect the skill documentation or output to mention optional paid customization/contact information.
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
Findings (1)
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)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The README presents all instructions, examples, and usage details only in Chinese. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly documented and justified, which is not present here.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The entire skill reference is written only in Chinese and does not indicate that users may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly documented and justified.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 85% confidence
- Finding
The description states "本技能面向中文会话设计", which sets a language/locale restriction in the skill's natural-language behavior. While later text notes non-Chinese conversations may have limited effectiveness, the file does not present a user choice or explicit opt-in for language handling.
- Content
Intent-Code Divergence
Low
- Category
- Not specified by scanner
- Confidence
- 78% confidence
- Finding
L14 states the conversations are handled entirely locally and not transmitted externally. However, the skill documentation also embeds an email-based contact path for feedback/customization, which introduces an external interaction channel not reflected in the '本地处理 + 脱敏' claim; this is a mild intent/documentation inconsistency rather than a code-level contradiction.
- Content