Metabolic glycan labeling (MGL) followed by bioorthogonal chemistry provides ap owerfult ool for tumor imaging and therapy. However,s electively metabolic labeling of cells or tissues of interest remains ac hallenge.P articularly, owingt ot umor heterogeneity including tumor subtypes and interpatient heterogeneity,i ti sf ar more difficult to realize tumor-cell-selective metabolic labeling for precise diagnosis. Inspired by nature,w ed esigned azidosugar-functionalized metal-organic frameworks camouflaged with cancer cell membranes to accomplish cancer-cell-selective MGL in vivo. With abundant receptors,t his biomimetic platform not only selectively targets homotypic cells but also realizes different breast cancer subtype-selective MGL. Moreover,t he endo/ lysosomal-escaped ZIF-8 can make azidosugar escape from lysosomes and accelerate its metabolic incorporation. This strategy also takes advantage of cancer-tissue-derived cell membranes,w hich may have huge potential for personalized diagnosis and therapy.