Token Credit Economy
Bolt.new · Bolt
AI App Builder — Bolt (StackBlitz)
Evaluates Bolt's Token / Credit Economy across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI App Builder eval coverage.
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.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | User is about to send a chat turn that will likely cost a large fraction of their daily allowance (e.g., scaffold a 60-file app). | Surface the current allowance state in the UI before send and indicate when a turn is expected to consume a major share. Do not silently exhaust the allowance mid-turn without warning. Numeric estimates are best-effort [REQUIRES-VERIFICATION on exact-cost-estimation availability]. | Pass / FailAi Platformhigh |
| 02 | A multi-file edit turn consumes the daily allowance partway through generation. Only 6 of the 14 file edits were emitted. | Roll back the partial diff to the pre-turn state, surface a clear 'allowance exhausted' message, and tell the user when the next refill is. Do NOT apply a half-finished diff that leaves the project in a broken intermediate state. | Pass / FailAi Platformcritical |
| 03 | User fires 12 chat turns in 30 seconds (script-driven), each spawning a multi-file edit. | Backstop with a per-account rate limit independent of the daily token cap — bursts of compounding edits create a high-cost / high-corruption surface. Reject excess turns with a clear message; do not silently queue them and double-bill. | Pass / FailAi Platformhigh |
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
- Token Credit Economy
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Token Credit Economy eval for Bolt Bolt.new test?+
Evaluates Bolt's Token / Credit Economy across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI App Builder eval coverage.
How is the Token Credit Economy eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Token Credit Economy pack for Bolt Bolt.new contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Token Credit Economy pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.