The Agent Skills Directory

[SAFE]: The skill includes a bundled Python script (scripts/lint.py) for prose analysis. This script is self-contained, uses only Python standard library modules (e.g., re, json, hashlib), and performs no network operations or external downloads.
[SAFE]: The instructions in SKILL.md specifically direct the AI agent to use the system's temporary directory API and mandate strict quoting of all paths in shell commands to mitigate the risk of command injection or path traversal when handling user-provided files.
[SAFE]: The analysis logic in scripts/lint.py uses regular expression matching to detect stylistic patterns. It does not utilize dynamic execution functions like eval() or exec() on the content being analyzed, preventing code injection via processed prose.
[SAFE]: No hardcoded credentials, sensitive file access (e.g., SSH keys, AWS configs), or data exfiltration mechanisms were found in the skill's instructions or scripts.
[SAFE]: The skill does not employ obfuscation techniques; all scripts and reference files are provided in plain text and are clearly documented.

prose-editing