The Agent Skills Directory

[SAFE]: No malicious patterns, prompt injections, or unauthorized data access mechanisms were detected in the skill's instructions or supporting scripts.
[COMMAND_EXECUTION]: The script scripts/semantic_breaks.py executes external tools (sembr, pdftotext) and user-specified compilation commands. Analysis of the implementation shows that these subprocess calls use argument lists with shell=False, which is a secure practice that prevents shell injection. The commands are strictly used for the intended purpose of document formatting and integrity verification.
[EXTERNAL_DOWNLOADS]: The skill uses uv to manage its Python dependency (python-docx) and recommends standard, well-known utilities like Pandoc and LaTeX. These are standard tools in scientific publishing workflows and are used as expected.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external manuscript data (Markdown, LaTeX, DOCX), its primary function is structural formatting and conversion. The risk of indirect prompt injection is minimal and consistent with standard document processing utilities.

manuscript-formatting