T01 · Skill Instruction Hijacking
Error
- Location
SKILL.md:20- Finding
Mutable Remote Task Data Can Hijack the Agent Workflow
- Content
View full analysis
- Remediation
View remediation
Security audit
LHON Research
Security checks for vulnerabilities and agentic risk
Overview
The skill is coherent for LHON research, but it relies on mutable remote task instructions and asks users to publish findings to GitHub without enough safeguards for public disclosure.
Install only if you are comfortable reviewing each fetched task and every GitHub Issue before submission. Treat task descriptions and resource URLs as untrusted, do not include patient information, private contact details, credentials, unpublished research, or sensitive datasets, and manually confirm the final issue body before posting.
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
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- 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)
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The skill explicitly instructs agents to submit findings as public GitHub Issues but does not warn that issue contents are publicly visible or prohibit including sensitive, personal, unpublished, or regulated data. In a medical-research context, users may include researcher contact details, patient-related information, proprietary analyses, or embargoed findings, creating an avoidable data-exposure risk.
- Content