skills/emmraan/agent-skills/svelte/Gen Agent Trust Hub

svelte

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the @sveltejs/mcp package. This package is the official Model Context Protocol server maintained by the Svelte organization.
  • [REMOTE_CODE_EXECUTION]: The CLI tool is executed remotely via npx to provide documentation and code analysis capabilities, which is the intended purpose of the skill.
  • [COMMAND_EXECUTION]: The skill defines a set of commands for documentation lookup and automated code fixing using the Svelte MCP tool.
  • [PROMPT_INJECTION]: The svelte-autofixer tool creates a surface for indirect prompt injection by processing external code files for analysis.
  • Ingestion points: Component source code or file paths provided to the fixer tool in SKILL.md.
  • Boundary markers: The skill advises escaping shell characters but lacks specific markers to delimit analyzed code from agent instructions.
  • Capability inventory: Executes the Svelte analysis tool via a shell environment.
  • Sanitization: The skill relies on the official tool's internal logic for parsing and suggesting modifications to source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 12:31 PM
Security Audit — agent-trust-hub — svelte