paper-spine-latex

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions and examples for executing standard shell commands to compile LaTeX documents and convert between formats. Specifically, it uses pdflatex, bibtex, and pandoc in SKILL.md and references/round4-template-integration.md. These are expected for a document processing skill.
  • [DATA_INGESTION]: The skill processes user-provided manuscript files (.tex, .bib, .docx) using both standard compilers and custom analysis scripts.
  • Ingestion points: Manuscript content is read by scripts/latex_guard.py, scripts/word_guard.py, and scripts/humanize_check.py to perform structural and linguistic analysis.
  • Boundary markers: No explicit boundary markers are used to separate user content from the agent's internal instructions.
  • Capability inventory: The agent is instructed to execute shell commands (pdflatex, bibtex, pandoc) and write various report files to the paper_rewriting_output directory.
  • Sanitization: Analysis scripts use regex patterns to identify structural errors but do not specifically sanitize content for hidden instructions, which is typical for developer-focused tooling.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 07:36 AM
Security Audit — agent-trust-hub — paper-spine-latex