BACKGROUND: People with disabilities frequently experience barriers in seeking healthcare that lead to poorer health outcomes compared to people without disabilities. To overcome this, it is important to assess the accessibility of primary health facilities - broadly defined to include a disability-inclusive service provision - so as to document present status and identify areas for improvement. We aimed to identify, adapt and pilot test an appropriate tool to assess the accessibility of primary health facilities in Luuka District, Uganda. METHODS: We conducted a rapid literature review to identify appropriate tools, selecting the Disability Awareness Checklist (DAC) on account of its relative brevity and development as a sensitization and action tool. We undertook three rounds of adaptation, working together with youth researchers (aged 18-35) with disabilities who then underwent 2 days of training as DAC facilitators. The adapted tool comprised 71 indicators across four domains and 12 sub-domains. We also developed a structured feedback form for facilitators to complete with healthcare workers. We calculated median accessibility scores overall, per domain and per sub-domain, and categorised feedback form suggestions by type and presumed investment level. We pilot-tested the adapted tool in 5 primary health facilities in one sub-district of Luuka, nested within a pilot healthcare worker training on disability. RESULTS: The median overall facility accessibility score was 17.8% (range 12.3-28.8). Facility scores were highest in the universal design and accessibility domain (25.8%, 22.6-41.9), followed by reasonable accommodation (20.0%, 6.7-33.3). Median scores for capacity of facility staff (6.67%, 6.7-20.0), and linkages to other services were lower (0.0%, 0-25.0). Within the feedback forms, there were a median of 21 suggestions (range 14-26) per facility. Most commonly, these were categorised as minor structural changes (20% of suggestions), with a third categorised as no (2%) or low (33%) cost, and the majority (40%) medium cost. CONCLUSIONS: Overall accessibility scores were low, with many opportunities for low-cost improvement at the facility level. We did not identify any issues with the implementation of the tool, suggesting few further adaptations are required for its future use in this setting.