The Agent Skills Directory

[SAFE]: No malicious instructions, obfuscation, or unauthorized data access patterns were identified in the skill files.- [NO_CODE]: The skill consists of instructional Markdown files and does not contain any executable scripts, binaries, or automated installation commands.- [DATA_EXFILTRATION]: No network activity or attempts to read sensitive system files (e.g., credentials or environment variables) were detected.- [COMMAND_EXECUTION]: The skill mentions command-line tools like pandoc for document conversion, but these are provided as informational examples for the user or agent rather than autonomous execution scripts.- [PROMPT_INJECTION]: The skill does not attempt to subvert the AI agent's constraints or safety guidelines. An analysis of the indirect prompt injection surface was performed:
Ingestion points: Processes user-supplied manuscript content (.md, .docx) and bibliography files (.bib) through conversion commands.
Boundary markers: Absent in the suggested pandoc conversion examples.
Capability inventory: Suggests file transformation using pandoc and text refinement using the manuscript:humanizer sub-skill.
Sanitization: None specified; the skill relies on the default behavior of the tools and the agent's reasoning during the formatting process.

manuscript-formatting