Tp2
High
- Category
- MCP Tool Poisoning
- Confidence
- 85% confidence
- Finding
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
- Content
Security audit
多智能体 3c4b57
Security checks for vulnerabilities and agentic risk
Overview
This is a simple multi-agent learning note skill with broad triggers but no executable code, credential access, persistence, or hidden behavior.
Installing this skill mainly adds low-assurance tutorial notes from video sources. Users should treat it as reference material and verify technical claims before relying on it, especially because its broad triggers may load it for general multi-agent questions.
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
- 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
Findings (3)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The manifest trigger contains broad marketing-style phrases and generic topic words such as '多智能体', making accidental activation likely for common multi-agent discussions. Because this skill is essentially learned content aggregated from videos rather than a tightly scoped operational tool, broad invocation increases the chance of inappropriate selection and unsafe reliance on low-assurance instructions.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The trigger conditions include broad phrases like '用户需要多智能体相关帮助', which can match ordinary user requests and cause unintended invocation. In an agent ecosystem, over-broad routing can expose users to irrelevant or untrusted skill behavior, increase prompt-surface area, and make downstream prompt-injection or unsafe action chains more likely.
- Content