The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation and instructional text. No executable scripts, command-line operations, or code blocks are present.
[DATA_EXPOSURE]: Analysis confirmed no sensitive file paths, hardcoded credentials, or network exfiltration patterns are present in the skill.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote downloads or execution of external code. Mentioned linting and formatting tools (e.g., eslint, prettier, black) are cited as best-practice examples for the user, not executed by the agent.
[PROMPT_INJECTION]: No malicious prompt injection patterns, role-play bypasses, or attempts to override safety guidelines were detected in the instructions.
[INDIRECT_PROMPT_INJECTION]: While the skill accepts user input via the $ARGUMENTS placeholder, it lacks any executable capabilities (file writes, network calls, or subprocess execution) that could be abused via an injection attack.

coding-standards