The Agent Skills Directory

[SAFE]: The skill's behavior is entirely consistent with its described purpose. It focuses on reading project documentation and writing configuration/rule files to standard locations like .claude/rules/ and CLAUDE.md.
[EXTERNAL_DOWNLOADS]: The skill references official Claude Code documentation (code.claude.com) and the author's public repository for reference and documentation gathering. These are trusted and well-known sources for the intended workflow.
[PROMPT_INJECTION]: The skill includes instructions for "Anti-Hallucination Guidelines" and validation steps, which are best practices for maintaining agent reliability and do not represent malicious injection or bypass attempts.
[DATA_EXFILTRATION]: No patterns of sensitive data access or external transmission were detected. The skill interacts with local project rule files which is necessary for its functionality.

create-rule