A key aspect of the practice of anaesthesia is the ability to perform practical procedures efficiently and safely. Decreased working hours during training, an increasing focus on patient safety, and greater accountability have resulted in a paradigm shift in medical education. The resulting international trend towards competency-based training demands robust methods of evaluation of all domains of learning. The assessment of procedural skills in anaesthesia is poor compared with other domains of learning and has fallen behind surgical fields. Logbooks and procedure lists are best suited to providing information regarding likely opportunities within training programmes. Retrospective global scoring and direct observation without specific criteria are unreliable. The current best evidence for a gold standard for assessment of procedural skills in anaesthesia consists of a combination of previously validated checklists and global rating scales, used prospectively by a trained observer, for a procedure performed in an actual patient. Future research should include core assessment parameters to ensure methodological rigor and facilitate robust comparisons with other studies: (i) reliability, (ii) validity, (iii) feasibility, (iv) cost-effectiveness, and (v) comprehensiveness with varying levels of difficulty. Simulation may become a key part of the future of formative and summative skills assessment in anaesthesia; however, research is required to develop and test simulators that are realistic enough to be suitable for use in high-stakes evaluation.