kestra-flow

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

• Third-party content exposure detected (high risk: 0.70). The skill's required Step 1 instructs the agent to fetch and directly read the live public schema from https://api.kestra.io/v1/plugins/schemas/flow, meaning third-party content is ingested and used to determine generation and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill explicitly instructs fetching the live schema at runtime from https://api.kestra.io/v1/plugins/schemas/flow and requires using that JSON to determine/validate allowed task types and properties, so the fetched content directly controls the agent's output generation.