paper-illustration-image2

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands and the Python subprocess module to manage its operations. The helper script scripts/paper_illustration_image2.py executes a codex binary to verify connectivity with the local app-server during the preflight stage.
  • [INDIRECT_PROMPT_INJECTION]: The skill's workflow ingests untrusted user data to plan and generate image prompts, creating a surface for indirect prompt injection.
  • Ingestion points: Step 1 in SKILL.md takes user requests, method notes, or paper sections as input.
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the user-provided data.
  • Capability inventory: The skill has access to file-writing capabilities, shell command execution (via the helper script and Bash tool), and several MCP tools for image generation and review.
  • Sanitization: There is no evidence of sanitization or validation performed on the user input before it is used to construct prompts for the agent or the image generation tool.
  • [DYNAMIC_EXECUTION]: The SKILL.md file contains logic for the agent to dynamically determine the path of the helper script paper_illustration_image2.py by searching through multiple possible installation locations (e.g., .agents/skills/, ~/.codex/skills/) before executing it using python3.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 09:43 AM