ml-paper-writing

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches metadata and conference style files from well-known academic services.
  • Instructions in SKILL.md use the requests library to fetch BibTeX metadata from doi.org.
  • The Makefile in templates/neurips2025/ uses curl to download official style packages from media.neurips.cc.
  • [COMMAND_EXECUTION]: The skill provides templates and logic for compiling LaTeX documents using standard tooling.
  • Compilation workflows described in templates/README.md and templates/neurips2025/Makefile execute pdflatex, bibtex, latexmk, and pdfcrop.
  • [INDIRECT_PROMPT_INJECTION]: The skill explores and processes content from research repositories, which represents an attack surface for indirect prompt injection.
  • Ingestion points: SKILL.md contains instructions for the agent to search and read contents from a local research repository (e.g., find, grep, xargs).
  • Boundary markers: The skill does not implement delimiters or explicit warnings to ignore embedded instructions when reading repository artifacts.
  • Capability inventory: The agent has network access (via requests and listed dependencies) and file system read access to perform its primary tasks.
  • Sanitization: No sanitization or filtering is performed on data read from the research repository before it is interpolated into the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 07:08 PM
Security Audit — agent-trust-hub — ml-paper-writing