The Agent Skills Directory

[SAFE]: The skill performs read-only analysis of project files (gtm-context.md, audit-report.json, package.json) and the general codebase to provide tracking recommendations.
[EXTERNAL_DOWNLOADS]: The skill mentions the installation of aimonk2025/google-tag-manager-automation via npx, which is a resource belonging to the skill's author. This is documented as a recommended setup step and is not executed automatically.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources including gtm-context.md, audit-report.json, and various codebase files found via scanning.
Ingestion points: File reads of gtm-context.md, audit-report.json, and codebase scanning using Glob and Grep tools.
Boundary markers: None explicitly defined in the instructions for delimiting external content.
Capability inventory: The skill is limited to analysis and console output; it does not perform network operations, file writes, or command execution beyond scanning.
Sanitization: No specific sanitization or validation of the ingested file content is mentioned.

gtm-quickstart