gtm-analytics

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements a data-ingestion workflow that is vulnerable to indirect prompt injection from local project files.
  • Ingestion points: The skill reads 16+ JSON files from the data/gtm/ directory and project-specific context from CLAUDE.md.
  • Boundary markers: No delimiters or isolation instructions are present to prevent the agent from following commands embedded within these files.
  • Capability inventory: The skill instructions are limited to data analysis and JSON file creation; no high-risk capabilities like shell commands (COMMAND_EXECUTION) or network requests (EXTERNAL_DOWNLOADS) were identified.
  • Sanitization: The skill does not define validation or sanitization logic for the content loaded from the project data files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 12:09 AM
Security Audit — agent-trust-hub — gtm-analytics