The Agent Skills Directory

[SAFE]: The skill uses official Google APIs (docs.googleapis.com and www.googleapis.com) for all document operations. No unauthorized domains or suspicious network activities were identified.
[SAFE]: Authentication is handled securely via standard OAuth 2.0 mechanisms, utilizing Zapier's managed connections or environment variables for tokens. No hardcoded credentials were found.
[SAFE]: Dependencies are restricted to standard, well-known libraries such as @zapier/connectors-sdk, @modelcontextprotocol/sdk, and zod.
[INDIRECT_PROMPT_INJECTION]: The skill possesses the surface for indirect prompt injection as it ingests data from external Google Docs.
Ingestion points: getDocument.ts, exportDocument.ts, findDocuments.ts, and findText.ts read content from document titles and bodies.
Boundary markers: None explicitly enforced by the skill code; the agent receives document content as structured data or text/markdown.
Capability inventory: A wide range of write operations across all editing scripts (e.g., appendText.ts, insertText.ts, deleteContentRange.ts) allow for document modification.
Sanitization: All tool inputs are strictly validated against Zod schemas.

google-docs