Back to skill

Security audit

Zynq Linux Hardware Debug

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Zynq hardware-debugging guide, but it includes high-impact hardware write examples without safety scoping or warnings.

Review before installing if the agent may run commands on real hardware. Use this only with operators who understand the target board, register map, bus numbers, and peripheral effects; require explicit approval before any write, FPGA load, kernel module change, or devmem/devmem2 command.

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
  • 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 (2)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill includes raw physical memory/register access examples using tools like devmem/devmem2 to read and write mapped addresses, including a concrete write value. These operations can immediately reconfigure FPGA/AXI peripherals, hang the bus, corrupt device state, or crash the system if the address or value is wrong, so presenting them without strong cautions is dangerous.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill provides direct hardware-write examples such as I2C register writes without any warning that these commands can alter device state, corrupt configuration, or interfere with attached peripherals. In a hardware-debugging context this is expected functionality, but omission of safety guidance still creates real risk because users may copy commands onto production or unknown hardware.

Content

No source excerpt is available for this finding.