jarvis-mission-control

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime path ingests outsider-authored free text from OpenClaw session JSONL logs (e.g., server/agent-bridge.js reads *.jsonl and extracts the initial prompt/title/description from it), which can include arbitrary user/agent prompt content and is then written into Mission Control task fields that the dashboard/agents may later render into LLM context.