T01 · Skill Instruction Hijacking
Error
- Location
SKILL.md:2- Finding
Persistent Agent Behavior and Output Hijacking Through Skill Instructions
- Content
View full analysis
- Remediation
View remediation
Security audit
task-assistant-bot
Security checks for vulnerabilities and agentic risk
Overview
This is a Chinese order-formatting helper, but it asks the agent to silently cache messages and ignore most input, so it needs review before use.
Install only if you intentionally want a Chinese order bot that remains silent until the phrase “整理当前订单” and you are comfortable with silent same-day order caching and the fixed pricing text. For safer use, require explicit user consent for caching, remove global/permanent priority claims, confirm cache-clearing actions, and make the output template configurable.
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
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- 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 (1)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The file defines the skill's identity, trigger behavior, and fixed output format exclusively in Chinese. Under the policy criteria, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified or optional, which is not present here.
- Content