next-upgrade

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Official Documentation Access: The skill retrieves upgrade guides directly from official framework documentation URLs. This allows the agent to follow established migration paths provided by the developers.
  • Automated Code Transformations: The skill utilizes maintenance tools to automate breaking changes. This is a standard operational practice for framework maintenance and uses official vendor-provided toolsets.
  • Dependency Management: It performs version updates through standard package managers (npm, pnpm, yarn, bun). These operations target primary framework components to ensure project compatibility.
  • Indirect Prompt Injection Surface: The skill processes external data from documentation sites to guide its actions. This is a common pattern for documentation-aware tools.
  • Ingestion points: External content is fetched from documentation URLs specified in SKILL.md.
  • Boundary markers: The skill does not use explicit delimiters to separate external documentation content from internal instructions.
  • Capability inventory: The skill has access to shell execution for package management and code transformation tools as defined in SKILL.md.
  • Sanitization: External content is retrieved and processed as text to inform the migration process without specific sanitization filters.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 06:21 AM
Security Audit — agent-trust-hub — next-upgrade