This is a web-server for predicting proteins secreted by Malarial Parasite P.falciparum into infected-erythrocyte. It is based on SVM method using amino acid composition module we have achieved maximum accuracy 83.2%. The method predicts the secretory proteins at the rate of 0.1% false prediction. The method is trained and tested on a dataset of 504 Plasmodium falciparum proteins (252 secretory & 252 non-secretory proteins) and evaluated using leave-one out cross-validation technique. The protein used for developing our method were obtained from PLASMODB, a database of plasmodium genome.