Org Ecosystem

Fail

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill recommends setting (setq org-confirm-babel-evaluate nil), which disables the standard security prompt in Emacs that requires user confirmation before executing code blocks. This configuration allows any code (shell, python, etc.) embedded in an Org-mode document to be executed automatically by the agent, representing a severe RCE risk if untrusted files are processed.
  • [COMMAND_EXECUTION]: The LaTeX export configuration patterns include the -shell-escape flag in (setq org-latex-pdf-process '("latexmk -pdf -shell-escape %f")). This flag allows the LaTeX compiler to execute arbitrary system shell commands, creating a direct vector for command injection during document generation.
  • [DATA_EXFILTRATION]: Automated scanners identified notes.org and journal.org as malicious domains. While these appear in the skill as local file path examples (e.g., ~/org/notes.org), the collision with blacklisted domains and the skill's reliance on external documents creates a risk profile where local data could be targeted or exfiltrated if the agent is confused by these resources.
Recommendations
  • AI detected serious security threats
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
HIGH
Analyzed
Jun 13, 2026, 12:34 AM
Security Audit — agent-trust-hub — Org Ecosystem