vercel-observability

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx @sentry/wizard to configure error tracking. Sentry is a well-known and trusted observability service.
  • [COMMAND_EXECUTION]: Provides commands for using the Vercel CLI (vercel logs) and curl to interact with Vercel APIs. These are standard operations for the stated purpose of the skill.
  • [INDIRECT_PROMPT_INJECTION]: Potential attack surface identified in Step 2 (Runtime Logs). If an attacker can inject malicious content into application logs (e.g., via request headers or body), an agent reading these logs via the provided commands could be exposed to indirect prompt injection. The skill includes a security note advising to exclude sensitive data and redact errors.
  • [SAFE]: References to the author's websites (jeremylongshore.com, intentsolutions.io, tonsofskills.com) are considered vendor resources and do not pose a security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:48 AM
Security Audit — agent-trust-hub — vercel-observability