Ae4
Medium
- Category
- analysis-evasion
- Confidence
- 80% confidence
- Finding
Suspicious Unicode normalization or mixed-script content
- Content
Security audit
Formula Decoder
Security checks for vulnerabilities and agentic risk
Overview
This is a coherent Chinese-language educational skill for explaining math and physics formulas, with no evidence of data access, commands, persistence, or hidden behavior.
Install this if you want a Chinese-language, highly structured teaching style for formulas and math/physics concepts. Be aware it may activate on broad educational phrases, but the inspected artifact does not show privileged access, local data handling, persistence, or network behavior.
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
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- 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 conditions are broad enough to activate on many ordinary educational or conceptual queries, causing the skill to intercept requests outside a narrowly bounded scope. Over-broad auto-invocation increases the attack surface for prompt collisions, unintended routing, and policy bypass-by-specialization, especially when the skill contains strong procedural instructions that may dominate the assistant’s response style.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The trigger list includes broad phrases such as '深度解析', '费曼式解释', and general math/physics explanation terms that can match many ordinary user requests outside a narrowly scoped skill boundary. This can cause the skill to activate unexpectedly, leading to prompt-routing ambiguity, unintentional takeover of general explanation tasks, and reduced user control over which behavior is invoked.
- Content
Vague Triggers
Low
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The initialization text directs immediate activation whenever a user enters any formula name or expression, without guardrails, exclusions, or a confidence threshold. This can force activation in ambiguous contexts and reduce the orchestrator’s ability to decide whether the skill is appropriate, creating unnecessary scope creep and possible interference with safer or more relevant skills.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 81% confidence
- Finding
The name, description, author field, and tags are written only in Chinese, and the manifest does not indicate that language selection is optional or user-controlled. Under the policy, forcing a specific language without opt-in can be a natural-language policy issue unless clearly documented as region-specific.
- Content
Static analysis
No suspicious patterns detected.