nextjs-devtools
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (MEDIUM): The skill uses
npx next-devtools-mcp@latestinscripts/start-server.shandSKILL.md. This pattern downloads and executes the latest version of a package from the public npm registry without version pinning or integrity checks, increasing susceptibility to supply chain attacks. - [REMOTE_CODE_EXECUTION] (MEDIUM): By executing code via
npx, the skill runs arbitrary third-party code on the user's machine with the user's local privileges. The package is not from a designated trusted source. - [PROMPT_INJECTION] (LOW): The skill has an Indirect Prompt Injection surface (Category 8).
- Ingestion points: Tools such as
list-routesandget-configingest local project metadata, including route names and configuration content. - Boundary markers: None identified in the wrapper scripts.
- Capability inventory: Execution of shell scripts and spawning of network-connected processes.
- Sanitization: No evidence of sanitization for project-sourced data is present in the provided scripts.
Audit Metadata