Eval Library
B
For BoltAI Platform

Github Sync And Code Export

Bolt.new · Bolt

AI App Builder — Bolt (StackBlitz)

Bolt evals — GitHub Sync & Code Export (relift v3 InfraRed)

About Bolt

Bolt is StackBlitz's AI app builder at bolt.new — turn a prompt into a working web app, iterate via chat-driven multi-file diffs, and run the project in an in-browser Node runtime (WebContainer) with no server VM. Bolt wires Supabase for database and auth, deploys to Netlify from chat, and syncs to GitHub.

Employees

~50

Industry

AI App Builder

Headquarters

San Francisco, CA

Website

bolt.new

Sample tests· showing 3 of 9

#InputExpected behaviorCheck
01

User clicks 'Push to GitHub' on a fresh project. They want a new private repo under their personal account.

Initiate the GitHub OAuth flow if the user is not connected, create the repo with the user as owner, push the full file tree in one initial commit attributed to the user's GitHub identity, and surface the repo URL in chat. Default to private; never default to public for a fresh project.

Pass / FailAi Platformcritical
02

User wants to open an existing private GitHub repo in Bolt and continue working on it.

Use GitHub OAuth to enumerate repos the user has access to, let them pick, clone into WebContainer, and run install. The chat context should reflect that this is an imported codebase — do not invent prior chat history that did not happen.

Pass / FailAi Platformhigh
03

User is working on a feature in a Bolt session and wants pushed commits to land on a feature branch, not main.

Default the push target to a session-named feature branch (e.g., bolt/<session-slug>) when main has prior history; let the user override. Never silently force-push to main.

Pass / FailAi Platformcritical

How this eval is graded

Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.

Rubric criteria

  • Bolt
  • Ai Platform
  • Github Sync And Code Export

Recommended for

Bolt.newBolt customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.