marp-slides
Warn
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill initiates the download and installation of additional presentation components from an external GitHub repository (
robonuggets/marp-slides). This repository is not identified as a pre-approved or trusted source. - [REMOTE_CODE_EXECUTION]: The installation workflow uses
npx skills addto fetch and integrate remote code into the agent's environment, which constitutes the execution of unverified third-party content. - [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands to check for the presence of the skill in global and project directories (
~/.agents/skills/marp-slides/SKILL.md). - [DYNAMIC_CONTEXT_INJECTION]: The documentation explicitly references the
!command syntax for Claude Code, instructing the agent to use it for automated environment verification and dependency installation. While used here for setup, this capability allows for direct shell interaction within the chat context.
Audit Metadata