moai-workflow-ci-loop

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

• Third-party content exposure detected (high risk: 0.75). Runtime path: the skill polls GitHub PR checks via gh pr checks and then ingests failure details/log URLs into the LLM context through the manager-quality spawn prompt (handoff JSON includes failedChecks[].logUrl and related failure text), which is outsider-authored content from GitHub/public CI logs rather than user-authored workspace text.