What problem does it solve? Choosing between chat, coding agents, models, and tool-enabled surfaces for a task is often done by intuition, leading to over-provisioned or under-capable setups. This Skill derives the required access, tools, model capability, reasoning effort, context, permissions, verification, and agent topology from the task itself and maps them only to confirmed available choices. ## Core Features & Use Cases - Independent dimension analysis: Evaluates access/tools, model capability, reasoning effort, context, permissions, verification, and topology separately instead of collapsing them into one light-to-heavy scale. - Explicit completion states: Returns one of Recommendation ready, Conditional recommendation, Setup change required, or Insufficient evidence, with prerequisites, unknowns, and the responsible next actor. - Safe topology selection: Chooses single session, sequential handoffs, or parallel agents only from already-defined work units, rejecting parallelism when shared state or file conflicts exist. - Use Case: Before starting an authorized production configuration rollout, ask which of two available setups should prepare the verified change; the Skill selects the one with repository access and deterministic validation rather than the one with merely stronger reasoning. ## Quick Start Ask the agent which available chat, coding agent, or tool-enabled setup should handle your concrete task, listing the choices you have confirmed.
