We present an accreditation protocol for the outputs of noisy intermediate-scale quantum computing devices. By testing entire circuits rather than individual gates, our accreditation protocol can provide an upper-bound on the variation distance between noisy and noiseless probability distribution of the outputs of the target quantum circuit of interest. Our accreditation protocol requires implementation of quantum circuits no larger than the target circuit, moreover it does not rely on brute-force classical simulations of the target circuit. It is therefore practical in the near term and scalable in the long term. Furthermore, our accreditation protocol relies on mild assumptions on the noise, as it only assumes that noise in single-qubit gates is bounded (but potentially gate-dependent) in diamond norm. We make no more assumptions on the noise in other parts of the circuit. Unlike all other characterisation, validation and verification techniques, our protocol is ready for implementation on present devices.