Repository LoreRepo Lore
The chronicle engine for open source

Turn GitHub repositories into stories.

Discover the evolution of any open-source project through narrative analysis. Paste a repository URL and read its history — chapter by chapter — alongside its timeline, DNA, and contributors.

Paste any public repository. Nothing is stored.

Summon
  • Generated in real time
  • No login · nothing stored
  • Any public repository
What it reads

Not a dashboard. A chronicle.

Analytics tell you what happened. Repository Lore tells you the story of how it happened.

Timeline reconstruction

Genesis, first release, development bursts, community growth and major milestones — placed in order and explained.

Repository DNA

Seven scores from 0–100 — Innovation, Stability, Community, Growth, Maintenance, Documentation and Testing — each with a plain-language rationale.

5 Storytelling Vibes

Experience the exact same lore as a docu, a fantasy epic, a sci-fi log, a corporate recap, or pure unhinged meme chaos.

Contributor roles

The Architect, the Feature Builder, the Bug Hunter, the Maintainer — roles inferred from real commit behaviour, with reasons.

Live visualisations

Commit activity, contributor ranking, release cadence, language distribution and growth — charted with Recharts.

Stateless by design

No accounts, no database, no tracking. Every report is computed on request from public data and then forgotten.

The ritual

From URL to legend in seconds

  1. 01

    Paste a repository

    Drop in any public GitHub URL — facebook/react, a full link, or even an SSH clone string.

    1
  2. 02

    We read GitHub

    Metadata, commits, contributors, releases, pull requests, issues and languages are fetched live from the GitHub API.

    2
  3. 03

    The history is analysed

    An engine detects milestones, scores the DNA, assigns contributor roles and reconstructs the timeline.

    3
  4. 04

    Read the lore

    The story is narrated across five chapters — switch narrative modes and explore the charts as you read.

    4
Open the codex

Stories already waiting to be read

A few legendary repositories to start with. Each one is generated fresh, the moment you open it.

JavaScript

facebook/react

A library that began as an internal experiment and reshaped how the web is built.

Read the lore
TypeScript

vercel/next.js

The framework that turned React into a full-stack platform, release by release.

Read the lore
TypeScript

microsoft/vscode

An editor born in a few months that grew into the workshop of millions.

Read the lore
Marginalia

Questions in the margins

The honest answers about how this works and what happens to your data.

No. There is no database and no account system. Each report is computed on request directly from the public GitHub API and held only long enough to render the page. Responses are cached briefly at the edge to be polite to the API, then forgotten.
By default, no — and that is deliberate. The lore is produced by a deterministic engine that maps real repository signals onto five hand-written story templates, so it is fast, free, private, and never invents facts. Optionally, if an AI key is configured, an "AI prose" toggle can rephrase that text in its chosen voice — still instructed never to change the underlying facts.
Any public GitHub repository with at least a little history. Private repositories are not supported (analysing them would require your credentials, which goes against the no-login, nothing-stored design).
Not as a visitor. If you self-host, an optional GITHUB_TOKEN environment variable raises the API rate limit from 60 to 5,000 requests/hour. It needs no scopes and is only ever read on the server.
The metadata, contributors, releases and languages are exact. Commit-based insights use the most recent window the API exposes (up to 100 commits), so velocity and role signals reflect recent activity. A few timeline placements (like community growth) are clearly marked as approximate.
Yes. Every report is also available from the stateless API route at /api/analyze?repo=owner/name, returning the full structured analysis you can build on.
Diagnostics

Check the live API status

Confirm GitHub connectivity, your rate limit, and whether AI prose is enabled — without spending any API quota.

GET /api/status

Press Check now to run the diagnostics.