mise-tasks

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly documents remote task inclusion (references/advanced.md "Remote Tasks": example include = ["https://example.com/tasks.toml"]), which instructs the agent to fetch and ingest arbitrary public URLs as task definitions—untrusted third‑party content that can change agent behavior and tool execution.