The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Google Ads API, such as search terms and campaign names, which could be used to influence agent behavior through adversarial content.
Ingestion points: The skill fetches search terms, campaign names, and ad group details via the GoogleAdsService.search method in scripts such as search_terms_report.py, list_campaigns.py, and get_keyword_performance.py.
Boundary markers: No explicit boundary markers or delimiters are used when presenting API-returned data to the agent. The output is formatted using tabulate without instructions to treat the content as untrusted data.
Capability inventory: The skill possesses significant write capabilities, including the ability to mutate campaigns, adjust budgets, and create ads via the CampaignService, CampaignBudgetService, and AdGroupAdService.
Sanitization: There is no evidence of sanitization or filtering of the text returned by the Google Ads API before it is rendered into the agent's context.

google-ads