latex-paper-en
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code or exfiltration patterns were found. Connectivity to
api.crossref.organdapi.semanticscholar.orgis used solely for legitimate bibliography metadata verification and qualifies as safe under trusted source rules. - [DYNAMIC_EXECUTION]: Local Python scripts execute TeX-related binaries to provide core functionality. These calls are restricted to known academic tools, and high-risk flags are guarded by user confirmation.
- [COMMAND_EXECUTION]: The skill uses
uvand standard TeX CLI tools in a controlled manner, with platform constraints further limiting execution scope. - [INDIRECT_PROMPT_INJECTION]: The skill acknowledges the risk of malicious content in source files (ingestion point:
tex_loader.py). It implements boundary markers in its instructions (SKILL.md) to prevent the agent from obeying instructions embedded in.texor.bibfiles. While it lacks sanitization of prose before analysis, its capabilities (exec/network) are restricted and require user oversight.
Audit Metadata