gdoc

The gdoc skill is a specialized tool designed for software agents operating within the SafeClaw environment to reliably fetch the full text content of publicly accessible Google Docs. By bypassing standard web-scraping utilities that often truncate or summarize content, this skill leverages low-level network requests to ensure that the entire document structure remains intact for analysis, summarization, or data extraction tasks. It is specifically built for research, content synthesis, and documentation workflows where data integrity is paramount.

• Converts Google Doc URLs by replacing standard edit suffixes with the mobilebasic view, which presents document content in a clean, plain-text-friendly format.

• Uses the curl command-line utility to perform direct HTTP GET requests, ensuring maximum compatibility and reliability within containerized environments.

• Redirects document output directly into temporary files, allowing for seamless integration with local read tools and subsequent processing workflows.

• Facilitates automated information retrieval from shared knowledge bases, public research papers, and collaborative drafts without requiring manual copy-pasting.

• To use this skill, ensure the document is set to public access via 'Anyone with the link' settings in Google Drive.

• The skill explicitly mandates the use of curl over tools like WebFetch to avoid data truncation issues commonly encountered with modern web-crawling libraries.

• The expected workflow involves a two-step process: initiating the document download to a temporary storage path, followed by triggering the local file reader to parse the extracted text.

• Users should be aware that this tool is strictly read-only and requires a valid URL path ending in /edit or a similar suffix to properly map the mobilebasic endpoint.

• Useful for automated research agents, documentation collectors, and parallelized information synthesis tasks within a sandboxed development environment.