Both for the welfare of the birds studied and the validity of the results obtained, it is important that biologging attachment methods should be suitably safe and effective. We describe a weak-link wing harness, designed for long-term attachment, which safely detaches from the bird without need for recapture, and a UK field trial on two species of gull with contrasting life-histories, Herring Gull Larus argentatus and Lesser Black-backed Gull L. fuscus. We fitted 15 GPS devices to Herring Gulls in 2014 using three different weak-link materials: cotton thread, cotton piping cord and nitrile rubber. Productivity and return rates were compared against those fitted with permanent harnesses and a control group. A further 36 weak-link harnesses were fitted to Lesser Black-backed Gulls in 2017 and 2018 using GPS-GSM devices that provided more accurate attachment durations. The weak-link design was suitable for Herring and Lesser Black-backed Gulls and harnesses detached completely. Devices fitted with cotton piping cord or nitrile weak-link harnesses remained attached to Herring Gulls for up to four years. Cotton thread was less durable, with harnesses detaching in 1–2 years. We did not detect any significant effects on breeding success or return rates from the harnesses, although small effects sizes cannot be discounted. Devices fitted to Lesser Black-backed Gulls with 6-ply cotton, 18-ply cotton or piping cord weak- link harnesses had respective mean (± standard error) minimum attachment durations of 102 ± 13.6 days (N = 15), 358 ± 42 days (N = 18) and 596 ± 225 days (N = 2). The use of weak-link harnesses can provide a flexible and favourable alternative to permanent designs mitigating some of the associated risks.