cove

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

• Third-party content exposure detected (high risk: 1.00). The cove-isolated.md workflow (Step 3: Sub-Agent Prompt Template and Task Tool Usage) explicitly instructs spawned sub-agents to use WebSearch, context7, Read, or other tools to research and cite sources from the open web, and the Reconciliation step requires those independent, external findings to override or revise the agent's answers, exposing the agent to untrusted third-party content that can influence actions.