ml-paper-writing

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and automates downloads from established academic and conference infrastructure.
  • LaTeX style files are fetched from the official NeurIPS domain (media.neurips.cc) via the provided Makefile.
  • Citation workflows utilize legitimate APIs for research metadata, including Semantic Scholar, CrossRef (api.crossref.org), and arXiv (export.arxiv.org).
  • LaTeX templates are linked to official GitHub repositories managed by established organizations such as acl-org, ICLR, and COLM-org.
  • [COMMAND_EXECUTION]: Provides standard instructions and scripts for academic workflows.
  • Includes shell commands for exploring research repositories (find, grep, ls) to identify contributions and existing citations.
  • Provides standard LaTeX compilation workflows using pdflatex, bibtex, and latexmk.
  • Recommends the installation of the Exa MCP tool for real-time academic search via a standard NPX command targeting the official exa.ai domain.
  • [SAFE]: Comprehensive review of the 47 files shows no signs of malicious intent or hidden behavior.
  • Instructions prioritize academic integrity by explicitly warning against hallucinating citations.
  • The logic is consistent with the primary purpose of assisting researchers in publication workflows.
  • No obfuscation, persistence mechanisms, or unauthorized credential access were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 03:16 AM
Security Audit — agent-trust-hub — ml-paper-writing