stackblitz-common-errors

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains explicit instructions to protect user privacy by redacting API keys, registry tokens, OAuth details, and cookies from diagnostic data.
  • [SAFE]: The skill defines clear boundaries for the agent, prohibiting the broad weakening of browser security controls (COOP, COEP, CSP) without explicit authorization.
  • [SAFE]: External network access via WebFetch is scoped to official documentation sources for the targeted technology (StackBlitz/WebContainers), which are well-known services.
  • [SAFE]: The skill implements a deterministic, evidence-based workflow that discourages speculative code changes or unsafe retries of dependency installations.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 09:26 AM
Security Audit — agent-trust-hub — stackblitz-common-errors