moai-platform-baas

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates data migration between various BaaS providers (such as Auth0, Clerk, and Supabase), creating an attack surface where untrusted data from source providers could influence agent behavior during processing or result in injection into the target system.
  • Ingestion points: API responses from provider management SDKs (e.g., Auth0ToClerkMigrator.migrate_users in examples.md).
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are present in the provided migration examples.
  • Capability inventory: The skill utilizes Read, Write, and Bash tools, alongside network access for provider APIs.
  • Sanitization: Example migration scripts demonstrate basic manual string escaping (e.g., SQL character replacement), which may not be sufficient for complex adversarial inputs.
  • [EXTERNAL_DOWNLOADS]: The skill documentation and examples reference the installation and use of official packages from well-known technology organizations, including Auth0, Clerk, Supabase, Neon, and Vercel. It also references the use of shared scripts from a consolidated repository using uv run.
  • [COMMAND_EXECUTION]: The skill provides instructions for standard development commands, such as npm install, docker build, and uv run, to manage project dependencies and deployments.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 12:32 PM
Security Audit — agent-trust-hub — moai-platform-baas