skills/brycewang-stanford/auto-empirical-research-skills/academic-paper-reviewer/Gen Agent Trust Hub
academic-paper-reviewer
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or dangerous commands were detected across the 18 files. The skill functions entirely through role-playing instructions and markdown templates.\n- [NO_CODE]: The skill does not include any Python, Node.js, or shell scripts. All logic is contained within agent persona definitions and structured templates.\n- [DATA_EXFILTRATION]: There are no network-capable commands (e.g., curl, wget) or external URL requests that would facilitate the exfiltration of user data.\n- [PROMPT_INJECTION]: The instructions are strictly focused on academic review personas. No patterns indicating bypass attempts, role-play jailbreaks, or system prompt extraction were found.\n- [DATA_EXPOSURE]: The skill processes academic papers provided by the user. It does not access system files, environment variables, or hardcoded credentials.
Audit Metadata