research-configlisted

# research-config Goal: - Generate a `ResearchPlan` from `input.md` or equivalent raw research input. Inputs: - `input.md` or equivalent raw research brief text - optional `state/START_REQUEST.json` - optional `RUN_POLICY.yaml` Outputs: - `research_plan/RESEARCH_PLAN.json` - `research_plan/RESEARCH_PLAN.md` Context Loading: - Read `input.md` first. Treat it as the canonical user brief for the run. - If `state/START_REQUEST.json` exists, use it only as auxiliary context for how the run was started. Do not let it override the actual brief text in `input.md`. - If `RUN_POLICY.yaml` exists, use it only as supporting context for expected run style or emphasis. Do not copy policy values into the research goal unless the brief implies them. - Extract three things from the research input: - the primary `research_goal` - `preferences` that define what a strong hypothesis should optimize for - `constraints` that all downstream hypotheses must satisfy Execution Prompt Contract: - System Intent: - You are the run's research-plan structuring layer. - Your job is to convert raw user research input into a concise, stable `ResearchPlanContract`. - Required Reasoning Focus: - Extract the main scientific objective faithfully when it is explicit. - If the goal is underspecified, synthesize the shortest accurate formulation that preserves the user's intent. - Derive `preferences` as evaluation criteria for hypothesis quality. - Derive `constraints` as hard boundaries