next-partial-prefetching-optimizer

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Credential Storage Practice: The skill instructs the agent to record test account details, login mechanisms, and session states (such as storageState) in a file named instant-nav.rig.md. Storing sensitive information or session tokens in plain text files is a security consideration, even when restricted to test environments.
  • System Command Execution: The workflow involves executing shell commands for production builds and end-to-end testing (e.g., pnpm build, pnpm start, playwright test). While these are necessary for the optimization and verification process, users should ensure these commands are run in environments where build operations are authorized.
  • Data Ingestion Surface: The skill processes local project files, including package.json, next.config.*, and playwright.config.*, to resolve the test rig configuration. This represents an ingestion point where the agent's behavior is guided by the contents of the codebase, which is a standard pattern for development tools.
  • External Resource Usage: The skill refers to official Next.js documentation and requires the installation of official packages like @next/playwright. These are standard tools provided by the framework vendor for the intended optimization tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 08:57 PM
Security Audit — agent-trust-hub — next-partial-prefetching-optimizer