mise-tasks

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

• Third-party content exposure detected (high risk: 0.90). This skill's advanced features explicitly allow importing remote task definitions via URLs (see references/advanced.md "Remote Tasks" with include = ["https://example.com/tasks.toml"]), which means the agent can fetch and ingest arbitrary public URLs containing user-provided task instructions that could directly influence execution and tool use.