pre-submission-report

Pass

Audited by Gen Agent Trust Hub on May 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands such as latexmk, mkdir, and grep to compile LaTeX documents and perform text-based integrity scans. These operations are scoped by the frontmatter configuration and are necessary for the skill's primary functionality.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from LaTeX files which could potentially contain indirect prompt injection instructions aimed at influencing the agent or the paper-critic sub-agent.
  • Ingestion points: Accesses .tex files within the project directory for placeholder scans and compilation.
  • Boundary markers: The skill does not employ specific boundary markers or 'ignore' instructions when reading the document content.
  • Capability inventory: The agent has access to restricted Bash execution, file modification tools (Write/Edit), and sub-agent/task invocation.
  • Sanitization: No content sanitization is performed on the data read from the LaTeX files before it is processed or passed to other tools.
Audit Metadata
Risk Level
SAFE
Analyzed
May 1, 2026, 06:19 PM