farm

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

• Third-party content exposure detected (high risk: 0.75). The required workflow dynamically selects and runs a “farmer” subagent from .claude/agents/$ARGUMENTS-farmer.md, and that subagent is explicitly intended to pull information from an external source into the vault, which would then be ingested into the LLM context as free text (indirect prompt-injection risk).