Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

Rest in Pull Requests: A Local Gemma Writes Eulogies for Abandoned Repos

Description and use

The Software Cemetery app analyzes public GitHub repositories, assigns a transparent vitality score, builds an evidence ledger, and generates a short eulogy using a local Gemma model, enabling users to assess and revive quiet projects responsibly.

Problem

Developers lack a respectful way to assess and document the status of quiet open‑source repositories, risking mislabeling healthy projects as abandoned.

Outcome

The tool provides a transparent assessment of repo vitality, a verifiable evidence ledger, and a locally generated eulogy that encourages respectful handling of quiet projects.

Workflow

  • Enter a public GitHub repo URL or owner/repo
  • The app calls GitHub REST API (5 calls) to fetch repo, commits, issues/PRs, releases, contributors
  • Compute features such as days since last commit, release age, contributor concentration
  • Apply transparent rules to calculate a vitality score 0‑100
  • Generate evidence ledger entries E1…En linked to GitHub pages
  • Prompt local Gemma 3 via Ollama to write a short eulogy with [E#] citations
  • Verify that every [E#] citation exists in the ledger
  • Export the case file as Markdown or view in browser

Evidence

The Software Cemetery gives those repositories a respectful send‑off instead. Paste in any public GitHub repository and it: collects the public traces the repo left behind: commits, issues and pull requests, releases and contributors; gives it a vitality score from 0 to 100 using rules you can read in a single file, with no black box; builds an evidence ledger, where every claim (E1, E2, …) links to the exact GitHub page that supports it; asks a local Gemma model to write a short, nostalgic, gently sarcastic life story with a citation after every factual sentence; and then checks those citations in code, so an invented [E99] gets flagged instead of trusted.

Source

The model is swappable: change OLLAMA_MODEL in .env and the whole app uses a different one.

Source

The score starts at 100 and subtracts for each warning sign. The rules live in one function, so anyone can audit the weights.

Source

  • Record date: 2026-11-01T13:28:46.504Z