vercel-sandbox
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for technical investigation and documentation. No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected.
- [EXTERNAL_DOWNLOADS]: The skill references official Vercel documentation (vercel.com) and the author's own GitHub repository (github.com/getsentry/junior). These are trusted or well-known services and do not represent a security risk in this context.
- [COMMAND_EXECUTION]: While the skill instructs the agent to inspect local files (e.g.,
packages/junior/src/chat/config.ts) and installed SDK types, these are standard static analysis operations for an integration-documentation skill and do not involve execution of untrusted code or arbitrary commands. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes external documentation. However, it implements clear steps for classification and establishes a hierarchy of truth (installed SDK first, then docs), which serves as a security best practice for minimizing the impact of potentially poisoned external content.
Audit Metadata