skills/hiyeshu/codeck/codeck-export/Gen Agent Trust Hub

codeck-export

Warn

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The script pptx/scripts/office/soffice.py performs runtime C compilation and process injection.
  • Evidence: The script hardcodes a C source string, writes it to a temporary file in the system temp directory, compiles it using gcc into a shared object library, and uses the LD_PRELOAD environment variable to inject the library into the soffice process.
  • [COMMAND_EXECUTION]: Several scripts execute system-level commands via subprocess.run for document processing and environment configuration.
  • Evidence: The skill executes gcc for shared library compilation, soffice for document conversion, pdftoppm for image generation, and git for diffing activities.
  • [EXTERNAL_DOWNLOADS]: The skill automates the installation of browser binaries from remote registries.
  • Evidence: SKILL.md contains the command npx playwright install chromium to fetch the browser required for PDF rendering.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by processing untrusted user-supplied project files with tools that have browser and shell capabilities.
  • Ingestion points: User project directory (HTML files).
  • Boundary markers: Absent; instructions do not include warnings to disregard instructions embedded in the input files.
  • Capability inventory: Shell command execution across multiple Python scripts and browser automation via Playwright.
  • Sanitization: Absent; HTML content is passed directly to rendering engines.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 2, 2026, 06:33 AM