sdd-linear

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

• Third-party content exposure detected (high risk: 0.75). The required workflow delegates all Linear reads/writes to the bundled linear-project-manager sub-agent, and the only LLM-ingested “free text” from Linear would be issue descriptions/comments fetched by that sub-agent (outsider-authored content from other users in Linear), which is then passed back into the orchestrator/LLM context as part of the state snapshot.