RepoBridge

Local-first GitHub repo inspector for grounded LLM context.

Diagnostics completed

karpathy/nanoGPT · run 36eec7b0-9f2d-4382-a1aa-902b04420262 · 2026-08-26 10:40 UTC

This run is partial. Every stage completed, but at least one was bounded — a limit, a budget, or a truncated listing. Each is quantified below.

8 info 1 partial 0 degraded 0 error

Stages

In pipeline order. A stage that never ran is absent rather than shown as healthy — for a run that failed early, that absence is the point.

validation 1 metadata 1 tree 1 classification 2 content 1 extraction 1 summarization 1 persistence 1

Level: info partial degraded error clear

Timeline

Showing 2 of 9 events in stage classification . The counts above always describe the whole run, never the filtered view.

  1. info classification 10:40:22

    Classified 32 entries: 24 selected, 2 skipped, 6 directories.

    selected
    24
    skipped
    2
    failed
    0
    directories
    6
    accounts_for_every_entry
    True
  2. partial classification 10:40:22

    2 entries were skipped without reading. Every one is listed with its reason in the file view.

    skipped_non_text_asset
    2

Diagnostics is an account of the run, not a second copy of it. Per-file detail for every entry — including everything skipped without reading — is in the file listing, which is why a repository with 22,924 skipped files still has a readable timeline here.