Title: Dwarf satellites of high-z Lyman Break Galaxies: a free lunch for JWST

Abstract: We show that the James Webb Space Telescope will be able to detect dwarf satellites of high-$z$ Lyman Break Galaxies (LBGs). To this aim, we use cosmological simulations following the evolution of a typical $M_\star\simeq10^{10}\rm M_\odot$ LBG up to $z\simeq6$, and analyse the observational properties of its five satellite dwarf galaxies ($10^7{\rm M_\odot}<M_\star<10^9{\rm M_\odot}$). Modelling their stellar emission and dust attenuation, we reconstruct their rest-frame UV-optical spectra for $6<z<6.5$. JWST/NIRCam synthetic images show that the satellites can be spatially resolved from their host, and their emission is detectable by planned deep surveys. Moreover, we build synthetic spectral energy distributions and colour-magnitude diagrams for the satellites. We conclude that the color $\rm F200W-F356W$ is a powerful diagnostic tool for understanding their physical properties once they have been identified. For example, $\rm F200W-F356W~\lesssim-0.25$ can be used to identify star-bursting ($\rm SFR\sim5~M_\odot yr^{-1}$), low-mass ($M_\star\lesssim5\times 10^8\rm M_\odot$) systems, with $\sim80\%$ of their stars being young and metal-poor ($\log(Z_\star/Z_\odot) < -0.5$).