qa-dashboard

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The SKILL.md ReportPortal setup explicitly instructs fetching and running a public docker-compose.yml from raw.githubusercontent.com ("curl -LO https://raw.githubusercontent.com/reportportal/reportportal/master/docker-compose.yml"), which is untrusted third-party content the workflow expects you to download and execute, so those external files could inject instructions that materially change tool behavior.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill includes a runtime command that downloads and runs remote orchestrator content—curl -LO https://raw.githubusercontent.com/reportportal/reportportal/master/docker-compose.yml followed by docker compose up -d—so it fetches external content that results in executing remote container images and is required for the ReportPortal setup.