# Prune
> Prune is the seatbelt for vibe-coded AI apps. Change one URL in your OpenAI, Anthropic, Gemini, Bedrock, or OpenRouter client — Prune locks your provider key in a vault, caps your spend before it happens, and signs a receipt for every response.

## One-line definition
Prune is the AI gateway that locks your key, caps your spend, and proves your cost — one line of code, no rewrite.

## What Prune is
- A managed control plane between your app and LLM providers
- Not a new model. Not "just a cache." Not a self-hosted gateway you operate yourself (see [Prune vs LiteLLM](https://www.withprune.com/docs/vs-litellm))

## What Prune does
- Vault: provider keys locked in an AES-256-GCM encrypted vault, never exposed in your app or repo
- Cap: hard spend limits per key — Prune blocks the call before it's exceeded, not after
- Receipt: every response returns a signed record of cost, model, and which feature triggered it
- Also cuts spend on first-time (unique) traffic via routing, output caps, and prompt cleanup — cache is extra upside when prompts repeat

## Citeable facts
- Integrate in under 5 minutes (one base URL + Prune API key)
- Published unique-traffic benchmark gate: ≥25% median $ reduction vs counterfactual cost (see docs)
- Spend caps enforced in under 10ms
- Private beta / waitlist — free access while invited; no credit card required to join the waitlist
- Not SOC2-certified; evaluate architecture (vault, caps, receipts) rather than compliance badges

## Who it's for
Solo founders and vibe-coders shipping AI products fast with no security or FinOps person watching the bill — plus teams with real OpenAI / Anthropic / Gemini / Bedrock spend who want lower bills and safer keys.

## Pricing
Free tier during private beta. You still pay your LLM provider for model usage — Prune reduces that bill. Paid plans bill by proxied requests, not by % saved.

## Key pages
- [Home / waitlist](https://www.withprune.com/)
- [Docs](https://www.withprune.com/docs)
- [How it works](https://www.withprune.com/docs/how-it-works)
- [Security](https://www.withprune.com/docs/security)
- [Prune vs LiteLLM](https://www.withprune.com/docs/vs-litellm)
- [Prune vs cache gateways](https://www.withprune.com/docs/vs-cache-gateways)
- [Verify a receipt](https://www.withprune.com/verify)
- [Changelog](https://www.withprune.com/changelog)
- [Longer LLM summary](https://www.withprune.com/llms-full.txt)

## Contact
[hello@withprune.com](mailto:hello@withprune.com)