ABSTRACT Standardised approaches to disability assessment or social security eligibility are not new. However, experiments in novel algorithmic approaches to assessment are beginning to appear. This article examines a proposed algorithmic assessment—dubbed ‘independent assessments’—under Australia’s National Disability Insurance Scheme (‘NDIS’), which sought to narrow and standardise eligibility criteria for the Scheme. The proposal was ultimately abandoned following sustained public backlash. Yet, disability assessment will continue to face change and contestation, making it timely to consider the abandoned proposal, including through a rights-based lens. The ‘independent assessment’ proposal appeared to violate the Convention on the Rights of Persons with Disabilities (‘CRPD’) on several fronts, including being overly standardised and impairment focused in ways that were contrary to the social model of disability that strongly influenced the CRPD. The assessment proposal lacked a clear review mechanism to appeal outcomes and relied on a narrow pool of outsourced professionals to undertake the assessment to the exclusion of input by the person’s own support specialists. Inattention to socio-cultural difference also raised concerns, particularly regarding Indigenous people with disabilities. This article concludes by asking what a rights-based disability assessment looks like, while critically examining the limits of a narrow focus on disability assessment methodology.