jfrog-init

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill contains legitimate setup instructions but explicitly directs storing and using access tokens (writing ~/.netrc and embedding token-based URLs) and other environment mutations that weaken local security boundaries, which is suspicious steering without clear malicious intent.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime install steps that fetch and execute remote code: the Node install uses curl "https://raw.githubusercontent.com/nvm-sh/nvm/${NVM_TAG}/install.sh" (piped to bash), the JFrog CLI installer can download binaries from "https://releases.jfrog.io/artifactory/jfrog-cli/v2-jf/[RELEASE]" and retries npm installs against "https://registry.npmjs.org/" — all of which are fetched/executed during runtime when the user accepts installation prompts.