scanned 7d ago·osv · gitleaks · opengrep · picklescan + heuristics·full breakdown in the Security tab ↓
E2B Fragments
E2B Fragments is a production-ready open-source starter that lets users generate and run full-stack web applications from natural language prompts. Built on Next.js, it uses E2B secure cloud sandboxes for safe code execution and streams generated apps in real time — think an open-source v0.dev or Replit Agent you can self-host.
Key features
AI code generation — Claude, GPT-4o, or Fireworks generate runnable apps from a text prompt
E2B sandbox execution — generated code runs in isolated cloud containers (no host machine risk)
Real-time streaming — code and UI appear progressively as the model generates
Multi-framework support — generates Next.js, Streamlit, Python, and Gradio apps
Bring your own model — swap in any supported LLM via a single env var
One-click deploy — ships with a Vercel deploy button and Docker support
Quick start
npx ai-supply add e2b-fragments-ai-app-starter
# Or clone and run directly
git clone https://github.com/e2b-dev/fragments
cd fragments
cp .env.example .env.local
# Set E2B_API_KEY and ANTHROPIC_API_KEY in .env.local
npm install
npm run dev
Open http://localhost:3000 and type a prompt like: "Build a to-do app with local storage".
Curated mirror of the open-source E2B Fragments project (Apache-2.0). Install upstream from the repository.
! Security: Review · 8888/100 · grade Bscanned 7d ago
✓ no compromise signals5 risk-surface · 6/20 OWASP controls flagged
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
OWASP Top 10 for LLM Applications
⚠LLM03Supply Chainhigh
Vulnerable/compromised dependencies, models or archives in the artifact.