skills/firzus/agent-skills/vite-plus/Gen Agent Trust Hub

vite-plus

Warn

Audited by Gen Agent Trust Hub on Oct 14, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The scripts/smoke.mjs file executes npm install, which downloads and installs third-party packages from the npm registry. The vp migrate and vp upgrade commands also facilitate the download of dependencies, package managers, and tool updates from external sources.\n- [EXTERNAL_DOWNLOADS]: The setup-and-migration.md file references the voidzero-dev/setup-vp GitHub repository for CI setup and directs users to fetch migration rules from https://viteplus.dev/guide/migrate-rules.\n- [COMMAND_EXECUTION]: The skill facilitates the execution of the vp CLI and npm commands. The smoke test script specifically uses spawnSync to execute build, test, and shell-based commands across different platforms.\n- [DYNAMIC_EXECUTION]: scripts/smoke.mjs dynamically generates source files, mock project manifests, and configuration scripts at runtime, subsequently executing them using Node.js with the -e (evaluate) flag.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data which creates a vulnerability surface for indirect instructions.\n
  • Ingestion points: Reads package.json, vite.config.*, tsconfig.json, and lockfiles to determine the project contract and tool configuration.\n
  • Boundary markers: None identified; the skill treats content within these configuration files as authoritative for its operations.\n
  • Capability inventory: Includes command execution (vp, npm), dynamic script execution, and file writing capabilities.\n
  • Sanitization: There is no documented mechanism to sanitize or validate the content ingested from external project manifests before using them to drive tool behavior.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 14, 2026, 06:07 PM