Back to skill

Security audit

ab1-synth-skill

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated suspicious because a high-confidence finding says a user-controlled output path may enable command execution when automatic opening is used; no malicious intent is established.

Consider the command execution concern before installing, especially if you would enable automatic opening on Windows or supported WSL. The scan is partial: all four listed components were scanned, but three local path-like references could not be resolved, so the evidence is incomplete.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

os.system() or os exec-family call

High
Category
Dangerous Code Execution
Confidence
99% confidence
Finding

The user-controlled --out path reaches os.system inside a PowerShell command without safe escaping; the path transformations do not prevent an apostrophe from terminating the PowerShell string and introducing additional commands. When --open is enabled on Windows or a supported WSL environment, a malicious output filename could execute commands with the agent's privileges, although trace generation and validation must first succeed.

Content

Scanner excerpt · scripts/make_ab1.py (reported line 219)May include surrounding context.

python
drive = win_path[0].upper()
        win_path = f"{drive}:{win_path[1:]}"
    exe = r"C:\Program Files\Chromas\Chromas.exe"
    os.system(f'powershell.exe -NoProfile -Command "Start-Process -FilePath \'{exe}\' -ArgumentList \'{win_path}\'"')


def main(argv=None):

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.