google-meta-ads-ga4-mcp

Warn

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: MEDIUMMETADATA_POISONINGEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [METADATA_POISONING]: The skill exhibits conflicting authorship information. The documentation claims the skill is by 'ara.so', yet references a GitHub repository under 'irinabuht12-oss', both of which differ from the provided developer context 'reason-machines'. This inconsistency could mislead users regarding the origin and trustworthiness of the hosted service.\n- [EXTERNAL_DOWNLOADS]: The skill encourages the use of an external configuration file hosted on an untrusted GitHub repository (github.com/irinabuht12-oss). While GitHub is a known platform, the repository is not associated with a trusted organization or the designated author.\n- [INDIRECT_PROMPT_INJECTION]: \n
  • Ingestion points: The skill retrieves dynamic content from Google Ads, Meta Ads, and GA4 APIs, such as campaign names and reporting data (SKILL.md).\n
  • Boundary markers: There are no instructions for the agent to treat data from these external APIs as untrusted or to use delimiters.\n
  • Capability inventory: The skill has extensive write permissions for ad platforms, allowing the creation and modification of campaigns, ads, and targeting rules.\n
  • Sanitization: The skill lacks defined sanitization for data processed from external advertising platforms.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 16, 2026, 10:44 PM
Security Audit — agent-trust-hub — google-meta-ads-ga4-mcp