felo-cli

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (SKILL.md and references/workflow.md) instructs calling the external Felo Chat API (POST https://openapi.felo.ai/v2/chat) to retrieve answers and included data.resources (link/title/snippet) from public web sources, so the agent will ingest and act on untrusted third-party content that could carry indirect instructions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly requires running remote code at runtime via the CLI command "npx -y @willh/felo-cli --json """ which downloads and executes the npm package @willh/felo-cli and thus can directly control prompt handling and execute remote code.