The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends connecting to cohnen/mcp-google-ads for automated GAQL access. This is a repository hosted by a third-party individual rather than a verified service provider.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of Search Terms Reports and account exports which could theoretically contain malicious instructions.
Ingestion points: Data entering the context via google-data.json, manual CSV/Excel exports, and GAQL query results through MCP tools.
Boundary markers: None identified. The instructions do not define delimiters or specific 'ignore embedded instructions' prompts for the ingested search data.
Capability inventory: The skill has the capability to write files to the local file system (e.g., google-audit-results.json, google-audit-results.md, GOOGLE-ADS-REPORT.md).
Sanitization: There are no documented sanitization steps for cleaning search term strings or metadata before they are processed by the agent.
[COMMAND_EXECUTION]: The documentation provides a shell command template (python3 scripts/api/google_fetch.py ...) for the user to generate data files. While intended for manual user execution, it relies on the presence of local scripts not fully defined in the main file.

ads-google