Hydrogen peroxide from a pharmacy bottle breaks down into water and oxygen, 2H2O2→2H2O+O2, so slowly that you see no bubbles. Drop in a pinch of black manganese dioxide powder and it foams at once. When the foaming stops, filter and dry the powder: it weighs what it did before, and it will do the same thing to a fresh sample. It sped the reaction up and came out unchanged.
How can something that ends up unchanged make a difference? It takes part along the way. The catalyst offers a different route from reactants to products, one with a lower activation-energy barrier. Reactant particles stick to its surface or react with it in an intermediate step, and a later step releases the catalyst again. With a lower barrier, a much larger share of collisions have enough energy, so the rate rises. On an energy diagram, the catalyzed path has a lower hump between the same start and end levels.
A catalyst is not just a helpful reactant. Magnesium dropped into hydrochloric acid also makes it fizz, but the magnesium is used up and turns into magnesium chloride, so it is a reactant. The test is whether the substance is recovered in the same amount. A catalyst also cannot change where an equilibrium settles: it speeds the forward and reverse reactions by the same factor, so you reach the same mixture sooner, not a mixture with more product.
Think of it like this. A catalyst is like a mountain guide who leads hikers along a lower pass between the same two valleys. More hikers cross each day, and the guide walks back to lead the next group. The guide cannot make the far valley any lower or make hikers stay there, just as a catalyst cannot change ΔH or the equilibrium mixture.
Core idea. A catalyst gives a reaction a different route with a lower activation-energy barrier, so more collisions succeed and the reaction runs faster. It takes part in intermediate steps but is regenerated, so it is not used up, does not change ΔH, and does not shift an equilibrium.