parallel▌

Delegate independent tasks to specialized agents working concurrently with isolated context. \n \n Dispatch one agent per problem domain when facing 3+ unrelated failures across different test files or subsystems \n Each agent receives focused scope, clear goal, and constraints—no inherited session context or history \n Agents work in parallel on independent investigations, then you review summaries and integrate non-conflicting fixes \n Best for timing issues, subsystem bugs, and exploratory fi

Coordinate parallel feature development with file ownership strategies, conflict avoidance rules, and integration patterns. \n \n Provides three file ownership models (by directory, module, or layer) and enforces the cardinal rule: one owner per file to prevent merge conflicts \n Defines interface contracts as read-only coordination points between implementers, allowing shared boundaries without direct file conflicts \n Covers two integration patterns (vertical slice for independent features, ho

Systematic debugging framework using competing hypotheses to identify root causes across multiple failure categories. \n \n Generates hypotheses across six failure mode categories: logic errors, data issues, state problems, integration failures, resource issues, and environment mismatches \n Establishes evidence standards with citation requirements (file:line references) and confidence levels (high/medium/low) to avoid confirmation bias \n Supports parallel agent investigation with structured re