presubmit-checks

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the make utility to verify document compilation. This results in the execution of shell commands defined in a local Makefile, which could be used to execute arbitrary code if the file is malicious.
  • [PROMPT_INJECTION]: The skill processes untrusted LaTeX and BibTeX files, creating an attack surface for indirect prompt injection.
  • Ingestion points: Reads main.tex, project .tex files, and .bib files.
  • Boundary markers: No specific delimiters are used to separate user data from instructions.
  • Capability inventory: The agent can execute local commands (make) and call other skills based on paper content.
  • Sanitization: The skill does not perform sanitization or validation of the input files before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 02:29 AM
Security Audit — agent-trust-hub — presubmit-checks