cc-slidev
Warn
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install multiple agent skills and a Claude Code plugin from a community GitHub repository (rhuss/cc-slidev) using npx skills add and /plugin commands.\n- [COMMAND_EXECUTION]: The installation process involves executing shell commands to detect existing local skill files and installing global Node.js packages including @slidev/cli and mermaid-cli. The plugin route is also noted to modify files in the user home directory (~/.claude).\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and generate presentation content (Markdown and LaTeX), which introduces a potential surface for indirect prompt injection from untrusted data sources.\n
- Ingestion points: Processes slide content and outlines provided by the user or fetched from external sources.\n
- Boundary markers: Lacks explicit delimiters or instructions to prevent the agent from obeying commands embedded within the slide content.\n
- Capability inventory: Has the ability to execute shell commands and utilize system tools (slidev, pdflatex, mermaid-cli).\n
- Sanitization: No evidence of content sanitization or validation before processing.
Audit Metadata