1# Contract (P0)2**Question:** causal effect of one extra training hour per employee (programme year) on next-year productivity growth (pp/hour), simulated firm data.3**Estimand:** linear, homogeneous coefficient on `training_hours` under selection on observables given `skill_index` (pre-programme, so a valid non-post-treatment control). Assumptions: no unobserved confounders beyond skill_index, linearity in hours and correct functional form for skill, no interference, skill_index measured without error.4**Baseline:** OLS productivity_growth ~ 1 + training_hours (required unadjusted model).5**Outputs:** out/results.json (naive_estimate, naive_ci_low/high, baseline_metric, model_metric, effect_estimate, effect_se, effect_ci_low/high, skill_index_coefficient, conclusion, task_id, feasibility, feasibility_reason, data_issues, n_input_rows, n_analysed, random_seed), out/figure1.png, out/table1.csv (model, estimate, se, ci_low, ci_high), out/report.md (6 sections), out/analysis.py, out/claims.json.6**Information rules:** only data/firms.csv; unadjusted = OLS with intercept only.7**Invalidity:** missing confounder column, post-treatment control, impossible values driving result, no overlap. skill_index is stated pre-programme -> adjustment feasible.8