single-slide-ppt

Warn

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses child_process.execSync to execute shell commands, specifically curl for downloading images and mkdir for workspace setup as described in SKILL.md and IMAGE_GUIDE.md.
  • [DYNAMIC_EXECUTION]: The core workflow involves generating JavaScript code at runtime and executing it using the Node.js runtime to create the presentation files.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download images from external URLs found during web searches and to install the pptx-automizer package from the public npm registry.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where image URLs are discovered via web search and then directly interpolated into a shell command. The lack of validation for these URLs could allow a malicious web page to trigger command injection.
  • Ingestion points: Image URLs are sourced from external web search results and page fetching.
  • Capability inventory: The skill utilizes execSync for shell execution and has file system write capabilities.
  • Boundary markers: No delimiters or instructions to ignore embedded instructions within fetched content are provided.
  • Sanitization: No logic is present to sanitize or escape the external URLs before they are executed in the shell.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 11, 2026, 05:42 PM