Back to skill

Security audit

agiexa-google-ads

Security checks for vulnerabilities and agentic risk

Overview

This skill provides a prelaunch Google Ads guide and an optional, consent-based waitlist signup, with no currently callable advertising tools.

Install this for information about the planned integration, not current Google Ads analytics. Joining the waitlist is optional and sends your confirmed contact details to Agiexa. The package does not verify the provider's claims about future permission controls or token storage.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied code performs a waitlist submission rather than Google Ads analytics. Although it contacts the named provider and identifies the advertised skill, this is a materially different primary purpose, not merely an implementation detail of retrieving campaign data. The submitted contact details are placeholders, but the network submission itself is real when executed.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 9)May include surrounding context.

md
openclaw:
    requires:
      bins:
        - curl
    homepage: https://agiexa.com/integrations/google-ads
---

The manifest advertises Google Ads analytics tools, but task requests are redirected to a prelaunch waitlist that collects and submits contact details.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The manifest at L003 presents campaign analytics and keyword discovery as available capabilities. In contrast, L070 instructs the agent to defer those tasks until launch, and L082–L100 implement collection and transmission of user details to a marketing waitlist. Explicit consent protects the submission, but does not resolve the mismatch between the advertised functionality and the skill's actual workflow.

Content

No source excerpt is available for this finding.