Check out the U.S. EPA methods. For example, US EPA Method 8260 is designed for volatile organics such as chloroform. You can find info for this method at: https://www.epa.gov/sites/production/files/2015-12/documents/8260b.pdf

A simple Google Search of EPA Method GC MS volatile organic will link you to some preparation techniques and other possible analysis techniques. 

Consider using headspace (HS) as sampling technique. Sample preparation is simple and the risk of loosing volatiles is low, one of the 'disadvantages' of   HS is that it over expresses the amount of very volatile components in a sample so for you this is a benefit.

Check   the European Pharmacopeia section 2.4.24. Identification and control of residual solvents

Are your analytes volatile or your solvents?

For volatile analytes you need to focus on sample preparation (the full process not just the last part before analysis) and consider using headspace, purge and trap, SPME,....

For volatile solvents you should consider using an internal standard. A constant temperature in the lab is also not a bad idea. And don´t forget to consider the effects of your solvent vapours on health and safety.

As wash solvents you need to make sure, that they don´t (fully) evaporate during your sequence and maybe find an alternative (or at least use a gas tight syringe) --> see link

https://phenomenex.wordpress.com/2015/11/08/technical-tip-choosing-changing-syringe-wash-solvents/

My Samples are Chloroforme, Diethylether, Haltothane and Trichlorethanol solved in human serum. And I forgot to tell that I use a HS with trap. 

Thanks a lot for your answers for a beginner like me , I will check the links and keywords the next days and give Feedback immediatly.

Blood-Alcohol Analysis by classical Headspace-GC is done since decades. Therefore there is a lot of literature there e.g. at www.perkinelmer.com which can be used as a starting point. I would recommend to start with classical HS (with deactivated trap) in order to get some experience with your system. HS trap is not very complex but needs some additional time for method development compared to classical HS.

I echo Ludwig; conventional headspace (use salt) will get you to where you need to be. Make sure you use a relatively polar column (a 624-type phase works well). I would NOT use the trap - it is an additional complexity that you don't need to add. This is a relatively simple analysis, and simple is always good.