Update to Marketing Swarm

{
  "name": "scenario-simulation",
  "description": "Use for campaign what-if questions, budget or KPI forecasts, sensitivity analysis, and decision ranges where deterministic or probabilistic scenarios are more useful than a single point estimate.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 285
    }
  ],
  "skill_md_contents": "---\nname: scenario-simulation\ndescription: Use for campaign what-if questions, budget or KPI forecasts, sensitivity analysis, and decision ranges where deterministic or probabilistic scenarios are more useful than a single point estimate.\n---\n\n# Scenario Simulation\n\nUse simulation to expose decision ranges and sensitivity. Do not decorate weak assumptions with false precision.\n\n## Choose the method\n\nUse the simplest defensible method:\n\n- deterministic scenario table for direct arithmetic changes\n- sensitivity analysis when one or more assumptions drive the result\n- bootstrap/resampling when representative historical observations are available\n- Monte Carlo only when probability distributions or defensible uncertainty ranges can be specified\n\nUse host-native Python when available for non-trivial calculations. Report the executed method and assumptions.\n\n## Define the model\n\nSpecify:\n\n- target metric and horizon\n- starting state\n- controllable inputs\n- uncertain inputs\n- constraints\n- relationship assumptions\n- number of simulations, if applicable\n\nDo not silently assume that CPA, ROAS, CVR, or CPM remains constant as spend changes. If a constant-rate scenario is useful as a baseline, label it explicitly.\n\n## Scenario set\n\nTypically compare:\n\n- base case\n- conservative case\n- expected/planning case\n- upside case\n\nFor budget decisions, include the current allocation as a control scenario.\n\n## Sensitivity\n\nIdentify which assumptions have the largest influence on the decision. If a small change in one uncertain parameter flips the recommendation, the answer should emphasize measurement rather than confidence.\n\n## Output contract\n\nReturn:\n\n- model and assumptions\n- scenario results or distributions\n- sensitivity drivers\n- decision boundary\n- what the model cannot infer\n- next measurement that would reduce uncertainty most\n\nA simulated result is not evidence that the future will occur. Keep observed data and modeled outcomes separate.\n"
}