start-core

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill documentation in SKILL.md and references/project-setup.md provides commands to scaffold projects using npx @tanstack/cli@latest and npx gitpick. These operations fetch resources from official and well-known framework repositories, which is standard for modern development workflows.
  • [DATA_EXPOSURE]: In references/common-mistakes.md, the skill proactively includes security guidance regarding the 'isomorphic' nature of TanStack Start loaders. It explicitly warns against including sensitive data like process.env.SECRET in client-accessible code and demonstrates the secure pattern of using createServerFn to encapsulate server-only logic.
  • [COMMAND_EXECUTION]: The agent workflow in SKILL.md specifies the use of fuse-ai-pilot:sniper for verification. This is an instruction to use an internal agent tool as part of a development lifecycle and does not represent a malicious execution pattern.
  • [SAFE]: The project templates and configuration files (package.json, tsconfig.json, vite.config.ts) follow industry best practices, including disabling verbatimModuleSyntax to prevent unintentional code leakage.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 06:13 AM