The Agent Skills Directory

[SAFE]: The skill performs legitimate developer-centric tasks such as generating templates and managing file organization. No malicious patterns or security vulnerabilities were identified in the instructions.
[COMMAND_EXECUTION]: The skill instructs the agent to perform file system operations, including reading project files, checking for directory existence, and writing new skill files to specific locations (~/.claude/skills/ or .claude/skills/). It also involves updating CLAUDE.md configuration files. These actions are transparent and aligned with the skill's primary purpose.
[PROMPT_INJECTION]: Analysis of the instructions reveals no attempts to override safety guardrails, extract system prompts, or subvert agent behavior. The process includes structured confirmation steps for format selection and file content before writing.
[DATA_EXFILTRATION]: No network operations, external downloads, or data transfer patterns to remote servers were found. The skill operates entirely on the local file system within the user's home or project directories.

skill-creator