The Agent Skills Directory

[SAFE]: The skill defines a legitimate instructional workflow for technical documentation. It utilizes standard platform commands such as /run, /how, and /to-artifact to gather context and generate visual narratives.
[INDIRECT_PROMPT_INJECTION]: The skill processes external information like PRs or issues to determine the subject of the documentation. While this is an ingestion point for untrusted data, the skill mitigates risk by dispatching a fresh, isolated agent to perform the final publishing step, ensuring a clean context for the generated artifact.
[COMMAND_EXECUTION]: The skill uses the /run command to start the application for screenshot purposes. This is an intended function for visual documentation and occurs within the environment's security boundaries.

showcase