A single jump filtration $({\mathscr{F}}_t)_{t\in \mathbb{R}_+}$ generated by a random variable $γ$ with values in $\overline{\mathbb{R}}_+$ on a probability space $(Ω,{\mathscr{F}},\mathsf{P})$ is defined as follows: a set $A\in {\mathscr{F}}$ belongs to ${\mathscr{F}}_t$ if $A\cap \{γ>t\}$ is either $\varnothing$ or $\{γ>t\}$. A process $M$ is proved to be a local martingale with respect to this filtration if and only if it has a representation $M_t=F(t){\mathbb{1}}_{\{t<γ\}}+L{\mathbb{1}}_{\{t\geqslant γ\}}$, where $F$ is a deterministic function and $L$ is a random variable such that $\mathsf{E}|M_t|<\infty$ and $\mathsf{E}(M_t)=\mathsf{E}(M_0)$ for every $t\in \{t\in \mathbb{R}_+:{\mathsf{P}}(γ\geqslant t)>0\}$. This result seems to be new even in a special case that has been studied in the literature, namely, where ${\mathscr{F}}$ is the smallest $σ$-field with respect to which $γ$ is measurable (and then the filtration is the smallest one with respect to which $γ$ is a stopping time). As a consequence, a full description of all local martingales is given and they are classified according to their global behaviour.