codex-visio-paper-figure-skill
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the user to clone a repository from an external GitHub account (
github.com/pengjunchi0/codex-visio-paper-figure-skill.git) to install the skill components. - [COMMAND_EXECUTION]: The skill relies on PowerShell scripts that are executed using the
-ExecutionPolicy Bypassflag, which disables local script execution restrictions to facilitate Visio and PowerPoint COM automation. - [INDIRECT_PROMPT_INJECTION]: The skill features an inherent attack surface for indirect prompt injection.
- Ingestion points: The skill ingests untrusted data in the form of reference images (screenshots, paper figures) and existing
.vsdxfiles (SKILL.md). - Boundary markers: No explicit boundary markers or instructions to ignore embedded textual instructions within the images are provided in the prompt logic.
- Capability inventory: The skill uses COM automation to control Visio and PowerPoint, allowing for file creation (
$docs.Add), file saving ($doc.SaveAs), and file exporting across multiple formats (SKILL.md). - Sanitization: While it includes coordinate validation helpers like
Assert-RelBoxto prevent geometric overlap, it lacks sanitization for text or data extracted from the reference images before using them in drawing commands.
Audit Metadata