ghbb - GitHub Backup Browser

ghbb is a local SQLite/FTS browser for archives produced by python-github-backup. It imports issue, pull request, discussion, release, comment/review, attachment manifest, and release asset metadata so humans and agents can search without GitHub API access.

Note: This project was 100% vibe coded. You can view the full agent session here. It took gpt-5.5:xhigh around 30 minutes and 10M tokens to produce it.

Quick start

uv sync
uv run ghbb import C:/CodeBlocks/ggml-org-backup/backup
uv run ghbb stats
uv run ghbb search "Fabrice Bellard"
uv run ghbb show ggml issue 1
uv run ghbb serve

Open http://127.0.0.1:8765/ for the web UI. JSON APIs are available under /api/stats, /api/repos, /api/search, and /api/items/{item_id}.

CLI examples

uv run ghbb search "KV cache" --repo llama.cpp --json
uv run ghbb search ggml_backend_cuda --repo ggml-org/llama.cpp --kind issue
uv run ghbb show llama.cpp pull 10001 --json
uv run ghbb show ggml discussion 32
uv run ghbb show llama.cpp release b1046

Database path resolution is intentionally simple:

1. GHBB_DB, when set and non-empty.
2. ./ghbb.db, relative to the current working directory.

There is no --db option in the MVP.

Import notes

• No GitHub API calls are made.
• Attachment files are not read into SQLite; only attachments/*/manifest.json metadata is imported.
• Re-running ghbb import rebuilds imported tables and the FTS index.
• Repository owner/name are derived from JSON URLs.

Planning and implementation details are in docs/.