modern-web-guidance

Fail

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions require the agent to run shell commands using npx to perform search and retrieval operations from a remote package.
  • [REMOTE_CODE_EXECUTION]: The use of npx -y modern-web-guidance@latest fetches and executes code directly from a public registry. Using the @latest tag is a high-risk behavior that introduces a supply chain vulnerability where an update to the package could execute arbitrary code on the user's system.
  • [EXTERNAL_DOWNLOADS]: Several implementation guides recommend dynamically importing scripts from external CDNs like unpkg.com, esm.sh, and jsdelivr.net, often with the @latest tag and without Subresource Integrity (SRI) hashes. Some suggested polyfills may proxy requests to remote servers, potentially compromising data privacy.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests and displays output from a remote command-line tool.
  • Ingestion points: Output from npx modern-web-guidance in SKILL.md.
  • Boundary markers: Absent.
  • Capability inventory: Shell command execution via npx.
  • Sanitization: Absent.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jun 22, 2026, 04:31 PM
Security Audit — agent-trust-hub — modern-web-guidance