RepoLens

Introduction

What RepoLens is, and why it leads with a verdict.

RepoLens is a Manifest V3 Chrome extension. Land on a GitHub, GitLab, npm, or PyPI page, click the toolbar icon, and it reads the repo, runs it past the AI provider of your choice, and opens a tab with a verdict-first breakdown.

Stars tell you a project is popular. They don't tell you whether it fits your problem. RepoLens answers the question you actually have: should I use this, and what am I signing up for?

The idea

Most repo pages are a sales pitch. The README leads with the best framing, the badges are curated, and the real trade-offs live three issues deep. RepoLens inverts that:

  • It opens with a verdict — a fit call (strong / solid / care / risky) and a one-line bottom line — so you know where you stand before you read a word of prose.
  • It measures instead of trusting — with the optional runner, Deep Dive is grounded in the actual source: real file counts, languages, the dependency graph, license, architecture, tests, and a secret scan.
  • It remembers, and helps you decide — every scan lands in a local library you can triage into Adopt / Trial / Hold / Reject, score against your own rubric, compare side-by-side, and even mine for new repos worth a look.

What it is not

  • Not a backend. There's no server, no account, no telemetry. Your keys and your library live on your machine.
  • Not a code executor. The optional runner downloads and reads source statically — it never runs repo code.

Where to go next

  • How it works — the whole picture, with diagrams: the scan pipeline, the three ways to connect a model, and the roadmap.
  • Getting started — install it and run your first scan.
  • The scan — the surfaces a scan opens into.
  • Triage & decide — keyboard-first decisions, a radar, collections, and drift alerts.
  • Evaluate & compare — score against your rubric, grade docs, and export a decision matrix.
  • Discover — search GitHub from inside, and recommendations from your own library.
  • Models — bring your own keys and route each part to a different model.
  • Storage — how your library is kept, and importing an old one.
  • The runner — measured facts for Deep Dive.