The processes taking place in the outermost reaches of spiral disks (the 'proto-disk') are intimately connected to the build-up of mass and angular momentum in galaxies. The thinness of spiral disks suggests that the activity is mostly quiescent and presumably this region is fed by cool flows coming into the halo from the intergalactic medium. While there is abundant evidence for the presence of a circumgalactic medium (CGM) around disk galaxies as traced by quasar absorption lines, it has been very difficult to connect this material to the outer gas disk. This has been a very difficult transition region to explore because baryon tracers are hard to observe. In particular, H I disks have been argued to truncate at a critical column density (N H > ∼ 10 19.5 cm −2 at 30 kpc for an L galaxy) where the gas is vulnerable to the external ionizing background. But new deep observations of nearby L spirals (e.g. Milky Way, NGC 2997) suggest that H I disks may extend much further than recognised to date, up to 60 kpc at N H ≈ 10 18 cm −2 . Motivated by these observations, here we show that a clumpy outer disk of dense clouds or cloudlets is potentially detectable to much larger radii and lower H I column densities than previously discussed. This extended proto-disk component is likely to explain some of the Mg II forest seen in quasar spectra as judged from absorption-line column densities and kinematics. We fully anticipate that the armada of new radio facilities and planned H I surveys coming online will detect this extreme outer disk (scree) material. We also propose a variant on the successful 'Dragonfly' technique to go after the very weak Hα signals expected in the proto-disk region.