The Agent Skills Directory

[SAFE]: The skill facilitates gathering context from external sources like Slack, Microsoft Teams, and cloud storage providers. This behavior is transparently explained to the user and requires their direction to fetch specific content, which is a secure approach to data ingestion.- [COMMAND_EXECUTION]: The skill uses standard agent tools such as create_file and str_replace to build and iterate on the document structure. These operations are restricted to the document being authored and are appropriate for the skill's stated purpose.- [SAFE]: The 'Reader Testing' phase uses sub-agents to verify document clarity. This is a self-correction mechanism and does not involve any dangerous operations or exposure of sensitive data.

doc-coauthoring