The Agent Skills Directory

[SAFE]: The skill consists of instructional text and code snippets for legitimate Google Sheets automation tasks. No malicious code, obfuscation, or suspicious network activities were identified. Analysis of the provided scripts shows they utilize standard Google Apps Script services (MailApp, UrlFetchApp) as intended for the described automation use cases.\n- [PROMPT_INJECTION]: The skill details workflows that ingest data from external sources like Google Forms. While this introduces a surface for indirect prompt injection at runtime, the documentation provides best practices for testing, and the skill itself contains no malicious instructions targeting the agent or system.\n- [DATA_EXFILTRATION]: The skill demonstrates how to use API keys and tokens through clearly marked placeholders such as YOUR_API_KEY and YOUR_TWITTER_TOKEN. This avoids the exposure of actual secrets while teaching necessary integration concepts. The use of network services is documented and consistent with the skill's purpose.

spreadsheet-automation