The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified. The skill is composed of natural language instructions focused on academic writing quality and logical consistency.
[SAFE]: The skill does not perform any network operations, external downloads, or exfiltration. All data access is limited to project-specific local paths required for paper polishing.
[SAFE]: No command execution, privilege escalation, or persistence mechanisms are present. The skill does not include any scripts or binary files.
[SAFE]: No obfuscation or prompt injection attempts were found in the text or metadata. The instructions are transparent and aligned with the skill's stated purpose.
[SAFE]: While the skill processes external input (paper drafts), it lacks dangerous capabilities like code execution or internet access that could be exploited via indirect prompt injection.

paper-polisher