zilliz

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

• Third-party content exposure detected (high risk: 0.70). The skill’s workflows and references (e.g., references/vector.md, references/monitoring.md, and SKILL.md) instruct the agent to run zilliz-cli commands that fetch and parse data from external sources—including user-uploaded collection/vector records and imports from S3 or cloud integrations—and explicitly recommend using --output json and formatting those results, so untrusted third-party content can be ingested and materially influence subsequent actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs running a remote install script at runtime with curl -fsSL https://raw.githubusercontent.com/zilliztech/zilliz-cli/master/install.sh | bash, which fetches and executes remote code and is presented as a required prerequisite for the skill.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly includes a "Billing" domain that handles "usage queries, invoices, payment methods" and even references a specific CLI command ("zilliz billing bind-card") for managing payment methods. Those are concrete billing/payment management operations (managing payment methods / billing), which constitute direct financial execution capability rather than a generic tool. Therefore it should be flagged.