The Agent Skills Directory

Command Execution: The skill uses common CLI utilities like gh, pac, and claude to collect system information and create GitHub issues. This is expected behavior for an issue-reporting tool and is limited to gathering relevant debugging data.
Data Handling and Disclosure: Because the skill posts to a public repository, it incorporates a clear warning and a verification step. This ensures that users have the opportunity to remove sensitive information like credentials or internal identifiers before submission.
Processing of Untrusted Input: The skill ingests user input from arguments and logs to populate the bug report. This creates a surface for potential indirect prompt injection where malicious text could influence the shell commands used to file the issue. This risk is addressed by the workflow's requirement for a human-in-the-loop preview and approval phase before the final command is executed.

report-issue