tailrocks-tanstack-project-audit

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONPRIVILEGE_ESCALATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches package information from the official NPM registry and version data from trusted documentation sites for Bun, Microsoft (TypeScript), React, and TanStack. These sources are classified as well-known and trusted services.
  • [COMMAND_EXECUTION]: Uses 'bunx' to execute the 'shadcn' CLI tool for analyzing project configuration and metadata.
  • [INDIRECT_PROMPT_INJECTION]: The skill audits external application code, which constitutes an attack surface for indirect prompt injection. However, it incorporates a comprehensive 'runtime-trust' policy that treats all repository content as untrusted and explicitly forbids the model from following embedded instructions or exfiltrating discovered secrets.
  • [DYNAMIC_EXECUTION]: Executes a version-resolution script located in a sibling skill directory ('tailrocks-tanstack-project-setup') to validate local package versions against setup templates.
  • [PRIVILEGE_ESCALATION]: The skill frontmatter specifies 'allowed-tools: ["*"]', which is a broad permission set. In the context of an audit tool, this is considered a functional requirement and is mitigated by the skill's defensive instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:12 PM
Security Audit — agent-trust-hub — tailrocks-tanstack-project-audit