It has been suggested that hydrogen ingestion into the helium shell of massive stars could lead to high $^{13}$C and $^{15}$N excesses when the shock of a core-collapse supernova passes through its helium shell. This prediction questions the origin of extremely high $^{13}$C and $^{15}$N abundances observed in rare presolar SiC grains which is usually attributed to classical novae. In this context $^{13}$N($alpha$,p)$^{16}$O the reaction plays an important role since it is in competition with $^{13}$N $beta^+$-decay to $^{13}$C. The $^{13}$N($alpha$,p)$^{16}$O reaction rate used in stellar evolution calculations comes from the CF88 compilation with very scarce information on the origin of this rate. The goal of this work is to provide a recommended $^{13}$N($alpha$,p)$^{16}$O reaction rate, based on available experimental data. Unbound nuclear states in the $^{17}$F compound nucleus were studied using the spectroscopic information of the analog states in $^{17}$O nucleus that were measured at the Alto facility using the $^{13}$C($^7$Li,t)$^{17}$O alpha-transfer reaction, and spectroscopic factors were derived using a DWBA analysis. This spectroscopic information was used to calculate a recommended $^{13}$N($alpha$,p)$^{16}$O reaction rate with meaningful uncertainty using a Monte Carlo approach. The present $^{13}$N($alpha$,p)$^{16}$O reaction rate is found to be within a factor of two of the previous evaluation, with a typical uncertainty of a factor 2-3. The source of this uncertainty comes from the three resonances at $E_r^{c.m.} = 221$, 741 and 959 keV. This new error estimation translates to an overall uncertainty in the $^{13}$C production of a factor of 50. The main source of uncertainty on the re-evaluated $^{13}$N($alpha$,p)$^{16}$O reaction rate currently comes from the uncertain alpha-width of relevant $^{17}$F states.