T02 · Agent Memory Poisoning
Warning
- Location
references/monitored_websites.md:19- Finding
Untrusted Remote Content Can Be Persisted into Skill Instructions
- Content
View full analysis
- Remediation
View remediation
Security audit
Cosmetics Compliance Manager
Security checks for vulnerabilities and agentic risk
Overview
The skill is a cosmetics compliance reference, but it documents automatic remote updates into the skill text without review or validation.
Review this skill before installing if you rely on it for compliance decisions. It should not automatically update its own skill text from monitored websites unless updates are reviewed, sourced from trusted HTTPS endpoints, and kept separate from executable agent instructions.
Vulnerability Patterns
- Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
- Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
- 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
- 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 (1)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The manifest description and heading present the skill entirely in Chinese and describe its usage only in that language, with no indication that users may choose another language. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified.
- Content