The Agent Skills Directory

[SAFE]: No security threats, malicious code, or deceptive patterns were identified. The skill's instructions are consistent with its stated purpose of guiding users through documentation creation.
[COMMAND_EXECUTION]: The skill uses expected file management tools such as create_file and str_replace to draft and edit sections of the document. These actions are user-guided and confined to the task of co-authoring.
[EXTERNAL_DOWNLOADS]: The workflow incorporates gathering context from external sources (e.g., Slack, Google Drive) via native platform integrations. These operations are initiated based on user input and are a primary feature of the document context-gathering process.
[PROMPT_INJECTION]: The skill processes external data (user-provided docs, links, and info-dumps), which is a characteristic of indirect prompt injection surfaces. However, as this is the core function of the co-authoring workflow and the agent operates under a structured sequence, it is considered safe.

doc-coauthoring