svelte-mcp

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the use of official Svelte tools by referencing @sveltejs/mcp and the remote endpoint https://mcp.svelte.dev/mcp. These sources are well-known and associated with the official Svelte framework maintenance.
  • [COMMAND_EXECUTION]: Setup instructions in references/documentation/local_setup.md include shell commands to install and run the MCP server using npx -y @sveltejs/mcp. These are standard procedures for initializing the Model Context Protocol environment.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection (Category 8) because it processes untrusted user-provided Svelte code for static analysis and documentation lookups.
  • Ingestion points: Svelte component code and developer queries are ingested via the svelte-autofixer tool and documentation tools defined in references/documentation/overview.md and references/documentation/prompts.md.
  • Boundary markers: The instructions do not provide explicit delimiters or 'ignore previous instructions' guards for the code snippets being analyzed.
  • Capability inventory: The skill has the ability to perform static analysis (svelte-autofixer) and generate external URLs (playground-link).
  • Sanitization: There is no documentation regarding the sanitization or escaping of the user-provided code before it is processed by the analysis tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 04:35 AM
Security Audit — agent-trust-hub — svelte-mcp