Dispose pure compounds in a separate bottle. With solutions it may be better to decompose or to precipitate the cyanide.
Just dispose as found in a separate bottle.
Maybe with solutions a treatment can save money.
a) Decomposing with hydrogen peroxide
The product of this reaction is ammonia/nitrogen and carbon dioxide. Therefore the resulting mixture may be simply flushed down the drain. But the reaction is only suitable for simple aqueous solutions. Since a lot of gas is formed during the reaction it is best to use a large beaker which is placed in a bowl, where the bowl is a safety measure if anything should bubble over. Adjust the pH to 10 - 11 and add excess hydrogen peroxide. The reaction is highly exothermic but needs an induction period. According to the literature the cyanide is oxidised to cyanate. But in own experiments forming of ammonia has been detected. When the developing of gas subsides change the pH to 8 - 9. Again a lot of gas is formed. Wait until no more gas and no more heat is developed. The Merck company suggests to check with a Merckoquant®Cyanide-teststick if all of the cyanide is decomposed. If you do not have this at hand at least you may check with a little crystal of iodide if there is an excess of oxidising agent. Verifying the cyanide as "Prussian Blue" generally will not work, since excessive peroxide will oxidise all iron(II) to iron(III), which will give poor soluble compounds.
b) Precipitation as "Prussian Blue"
This method is easy. It works with aqueous solutions and also in solutions containing organic solvents like ethylene glycol. Simply add a solution of iron(II)-sufate. Generally the iron(III) which is needed for the "Prussian Blue" is formed by autoxidation when the iron(II) has contact with air. Therefore the formed precipitate changes the color gradually from red to blue. You may accelerate this reaction by passing air through the solution. Depending om the Fe2+/Fe3+-ratio different products may be formed. The precipitate is filtered, washed with water and then disposed in a separate bottle. Subsequent oxidising according to a) does not work.