research-paper-writing
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches LaTeX style definitions from the official NeurIPS domain (
media.neurips.cc) and retrieves academic metadata from trusted providers including Semantic Scholar, CrossRef, and arXiv. These operations are essential for the skill's purpose and use reputable sources. - [COMMAND_EXECUTION]: Executes shell commands for LaTeX compilation, version control with Git, and background process management for long-running experiments. This includes the use of
latexmk,pdflatex, andpsfor legitimate monitoring. - [INDIRECT_PROMPT_INJECTION]: The pipeline ingests external text from academic sources via web tools. The risk of indirect injection is mitigated by structured verification workflows (cross-referencing citations) and explicit instructions to the agent to treat external content as unverified.
- [DATA_EXFILTRATION]: No exfiltration of sensitive information or unauthorized network communication was detected. The skill includes specific guidance for the user to avoid including credentials or personal data in repositories.
- [CREDENTIALS_UNSAFE]: No hardcoded API keys or secrets are present in the skill files. The instructions recommend standard practices for local secret management when using academic APIs.
Audit Metadata