diff --git a/skills/cuopt-numerical-optimization-api-c/SKILL.md b/skills/cuopt-numerical-optimization-api-c/SKILL.md
index b25acc14ba..9362936b88 100644
--- a/skills/cuopt-numerical-optimization-api-c/SKILL.md
+++ b/skills/cuopt-numerical-optimization-api-c/SKILL.md
@@ -35,6 +35,10 @@ QP uses the same library, include/lib paths, and build pattern as LP/MILP — on
 - **Continuous variables only** — set `CUOPT_CONTINUOUS` for every variable; integer QP is not supported.
 - **Q should be PSD** for a convex problem.
 
+## Dual values (LP / QP)
+
+`cuOptGetDualSolution` and `cuOptGetReducedCosts` return duals and reduced costs for **LP and QP**. They are not returned for a problem with quadratic constraints (the arrays are filled with `NaN`), so read them only when all constraints are linear. See [assets/lp_duals](assets/lp_duals/) for the call sequence.
+
 ## Debugging (MPS / C)
 
 **MPS parsing:** Required sections in order: NAME, ROWS, COLUMNS, RHS, (optional) BOUNDS, ENDATA. Integer markers: `'MARKER'`, `'INTORG'`, `'INTEND'`.
diff --git a/skills/cuopt-numerical-optimization-api-python/SKILL.md b/skills/cuopt-numerical-optimization-api-python/SKILL.md
index a7460c7c7c..87d3d247f9 100644
--- a/skills/cuopt-numerical-optimization-api-python/SKILL.md
+++ b/skills/cuopt-numerical-optimization-api-python/SKILL.md
@@ -253,13 +253,14 @@ settings.set_parameter("log_to_console", 1)
 | QP rejected with MAXIMIZE | QP only supports MINIMIZE | Negate the objective: minimize `-f(x)` |
 | QP returns non-optimal | Q not PSD or variables badly scaled | Check Q is PSD; rescale variables to similar magnitudes |
 
-## Getting Dual Values (LP only)
+## Getting Dual Values (LP / QP)
+
+Duals and reduced costs are returned for **LP and QP**. They are not returned for a problem with quadratic constraints (every value comes back as `NaN`), so read them only when all constraints are linear. MILP returns no duals.
 
 ```python
 if problem.Status.name == "Optimal":
-    constraint = problem.getConstraint("resource_a")
-    shadow_price = constraint.DualValue
-    print(f"Shadow price: {shadow_price}")
+    constraint = problem.getConstraint("resource_a")   # linear constraint
+    print(f"Dual value: {constraint.DualValue}")       # NaN if the model has quadratic constraints
 ```
 
 ## Reference Models