The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run and bash scripts to execute quarto render, bibtex-tidy, and various file operations. These executions are necessary for the skill's function of assembling and rendering research documents and do not exhibit malicious intent.\n- [EXTERNAL_DOWNLOADS]: The scripts/verify_doi.py script connects to the Crossref API (api.crossref.org) to validate and find DOI metadata for the manuscript's bibliography. This is a legitimate use of a well-known scientific service.\n- [PROMPT_INJECTION]: The skill processes external data (e.g., search results, extraction tables) to populate manuscript templates, creating a surface for indirect prompt injection. However, it implements boundary markers (e.g., ) and is designed for research contexts where data is generated by the user's own pipeline, resulting in a low risk.\n- [SAFE]: No signs of credential exfiltration, unauthorized persistence, or privilege escalation were detected. The skill's architecture relies on local project file manipulation and standard research software, which is typical for academic automation tools.

ma-manuscript-quarto