Back to skill

Security audit

google-ads-copy-generator

Security checks for vulnerabilities and agentic risk

Overview

The verdict is suspicious with medium confidence because the scanner reports a mismatch between the skill’s advertised ad-copy generation and its actual checking behavior, alongside an undeclared file-writing capability; no malicious intent is established.

Consider the suspicious Review verdict before installing. The reported behavior may not match the advertised purpose, validation may be unreliable, and file-writing scope is undeclared. The scan was partial because some referenced paths were missing, so it does not provide complete evidence about those references.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The primary advertised capability—ad-copy generation—is absent. The output contains field status and character counts rather than generated copy. The validation also does not fulfill the promise to check every field against real limits: lists are measured as concatenated strings, and Performance Max-specific text fields are not covered. Local input reading and report writing are ordinary supporting operations; no unrelated network access or account modification is present.

Content

No source excerpt is available for this finding.

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

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.

Character-limit validation checks combined asset lists rather than individual ad assets.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The manifest promises checks against real Google Ads character limits, which apply to each headline or description. At L092–L099, the code joins all entries with '; ' and checks the combined length against a single asset's limit, so two valid 20-character headlines are incorrectly reported as exceeding the 30-character limit.

Content

No source excerpt is available for this finding.