agentleFS
Sign inStart free
StevenicStevenic/vectra/CLAUDE.md

vectra

CLAUDE.md633 starsChanged 7 months agoMITView on GitHub

A lightweight, file-backed vector database for Node.js and browsers with Pinecone-compatible filtering and hybrid BM25 search.

We split this CLAUDE.md into the rules it gives an agent

5 rules. About 0 are common, 0 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run npm run test:mocha for fast iteration (skips rebuild)

    line 39
  2. Very specific:

    Run yarn lint and fix any errors

    line 50
  3. Very specific:

    Run yarn test and ensure all tests pass

    line 51
  4. Very specific:

    Update llms.txt if you changed public API surface

    line 52
  5. Very specific:

    Update docs/changelog.md with a summary of changes

    line 53

≈ “Common” is an estimate: at least 1 in 5 of the 25,680 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Vectra

A lightweight, file-backed vector database for Node.js and browsers with Pinecone-compatible filtering and hybrid BM25 search.

Project Structure

  • src/ — TypeScript source files
  • lib/ — Compiled output (do not edit directly)
  • bin/ — CLI entry point (vectra)
  • docs/ — Jekyll documentation site (GitHub Pages via just-the-docs)
  • samples/ — Runnable example projects
  • dist/ — Browser bundle (do not edit directly)

Development Commands

  • yarn install — Install dependencies
  • yarn build or npm run build — Compile TypeScript + copy proto schemas and templates
  • yarn build:browser — Build browser UMD bundle via webpack
  • yarn build:all — Build both Node.js and browser targets
  • yarn test or npm test — Build + run tests
  • npm run test:mocha — Run tests only (skip build)
  • yarn lint or npm run lint — Run ESLint
  • yarn clean — Remove build artifacts

Code Conventions

  • TypeScript strict mode — no any unless unavoidable
  • Tests colocated with source: src/foo.spec.ts tests src/foo.ts
  • Barrel exports through src/index.ts — every public API must be re-exported here
  • Browser exports through src/browser.ts — excludes Node-specific modules
  • Internal utilities in src/internals/ — not exported from the package
  • Proto schemas live in src/codecs/schemas/*.proto and are copied to lib/ at build time
  • Language binding templates in src/templates/ are copied to lib/templates/ at build time

Testing

  • Framework: mocha with sinon for mocks
  • Coverage: nyc — reports to Coveralls
  • Run npm run test:mocha for fast iteration (skips rebuild)
  • Test timeout: 10000ms

CI

  • GitHub Actions on push/PR to main
  • Pipeline: install → lint → build → test → coverage upload
  • Node 22 on ubuntu-latest

Before Submitting a PR

  • Run yarn lint and fix any errors
  • Run yarn test and ensure all tests pass
  • Update llms.txt if you changed public API surface
  • Update docs/changelog.md with a summary of changes

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free