Open Source AI Artifacts and Code Execution

Fragments by E2B

This is an open-source version of apps like Anthropic's Claude Artifacts, Vercel v0, or GPT Engineer.

Powered by the E2B SDK.

→ Try on fragments.e2b.dev

Features

• Based on Next.js 14 (App Router, Server Actions), shadcn/ui, TailwindCSS, Vercel AI SDK.
• Uses the E2B SDK by E2B to securely execute code generated by AI.
• Streaming in the UI.
• Can install and use any package from npm, pip.
• Supported stacks (add your own):
  • 🔸 Python interpreter
  • 🔸 Next.js
  • 🔸 Vue.js
  • 🔸 Streamlit
  • 🔸 Gradio
• Supported LLM Providers (add your own):
  • 🔸 OpenAI
  • 🔸 Anthropic
  • 🔸 Google AI
  • 🔸 Mistral
  • 🔸 Groq
  • 🔸 Fireworks
  • 🔸 Together AI
  • 🔸 Ollama
• Integrates with Morph Apply model for token efficient, accurate and faster code editing.

Make sure to give us a star!

Get started

Prerequisites

• git
• Recent version of Node.js and npm package manager
• E2B API Key
• LLM Provider API Key

1. Clone the repository

In your terminal:

1
git clone https://github.com/e2b-dev/fragments.git

2. Install the dependencies

Enter the repository:

1
cd fragments

Run the following to install the required dependencies:

1
npm i

3. Set the environment variables

Create a .env.local file and set the following:

1
# Get your API key here - https://e2b.dev/
2
E2B_API_KEY="your-e2b-api-key"
3

4
# OpenAI API Key
5
OPENAI_API_KEY=
6

7
# Other providers
8
ANTHROPIC_API_KEY=
9
GROQ_API_KEY=
10
FIREWORKS_API_KEY=
11
TOGETHER_API_KEY=
12
GOOGLE_AI_API_KEY=
13
GOOGLE_VERTEX_CREDENTIALS=
14
MISTRAL_API_KEY=
15
XAI_API_KEY=
16

17
### Optional env vars
18

19
# (on by default) Get your MORPH key here - https://morphllm.com/dashboard/api-keys
20
MORPH_API_KEY=
21

22
# Domain of the site
23
NEXT_PUBLIC_SITE_URL=
24

25
# Rate limit
26
RATE_LIMIT_MAX_REQUESTS=
27
RATE_LIMIT_WINDOW=
28

29
# Vercel/Upstash KV (short URLs, rate limiting)
30
KV_REST_API_URL=
31
KV_REST_API_TOKEN=
32

33
# Supabase (auth)
34
SUPABASE_URL=
35
SUPABASE_ANON_KEY=
36

37
# PostHog (analytics)
38
NEXT_PUBLIC_POSTHOG_KEY=
39
NEXT_PUBLIC_POSTHOG_HOST=
40

41
### Disabling functionality (when uncommented)
42

43
# Disable API key and base URL input in the chat
44
# NEXT_PUBLIC_NO_API_KEY_INPUT=
45
# NEXT_PUBLIC_NO_BASE_URL_INPUT=
46

47
# Hide local models from the list of available models
48
# NEXT_PUBLIC_HIDE_LOCAL_MODELS=

4. Start the development server

1
npm run dev

5. Build the web app

1
npm run build

Customize

Adding custom personas

1. Make sure E2B CLI is installed and you're logged in.

2. Add a new folder under sandbox-templates/ [blocked]

3. Initialize a new template using E2B CLI:

   1
   e2b template init

   This will create a new file called e2b.Dockerfile.

4. Adjust the e2b.Dockerfile

   Here's an example streamlit template:

   1
   # You can use most Debian-based base images
   2
   FROM python:3.19-slim
   3
   
   4
   RUN pip3 install --no-cache-dir streamlit pandas numpy matplotlib requests seaborn plotly
   5
   
   6
   # Copy the code to the container
   7
   WORKDIR /home/user
   8
   COPY . /home/user

5. Specify a custom start command in e2b.toml:

   1
   start_cmd = "cd /home/user && streamlit run app.py"

6. Deploy the template with the E2B CLI

   1
   e2b template build --name <template-name>

   After the build has finished, you should get the following message:

   1
   ✅ Building sandbox template <template-id> <template-name> finished.

7. Open lib/templates.json [blocked] in your code editor.

   Add your new template to the list. Here's an example for Streamlit:

   1
   "streamlit-developer": {
   2
     "name": "Streamlit developer",
   3
     "lib": [
   4
       "streamlit",
   5
       "pandas",
   6
       "numpy",
   7
       "matplotlib",
   8
       "requests",
   9
       "seaborn",
   10
       "plotly"
   11
     ],
   12
     "file": "app.py",
   13
     "instructions": "A streamlit app that reloads automatically.",
   14
     "port": 8501 // can be null
   15
   },

   Provide a template id (as key), name, list of dependencies, entrypoint and a port (optional). You can also add additional instructions that will be given to the LLM.

8. Optionally, add a new logo under public/thirdparty/templates [blocked]

Adding custom LLM models

1. Open lib/models.json [blocked] in your code editor.

2. Add a new entry to the models list:

   1
   {
   2
     "id": "mistral-large",
   3
     "name": "Mistral Large",
   4
     "provider": "Ollama",
   5
     "providerId": "ollama"
   6
   }

   Where id is the model id, name is the model name (visible in the UI), provider is the provider name and providerId is the provider tag (see adding providers below).

Adding custom LLM providers

1. Open lib/models.ts [blocked] in your code editor.

2. Add a new entry to the providerConfigs list:

   Example for fireworks:

   1
   fireworks: () => createOpenAI({ apiKey: apiKey || process.env.FIREWORKS_API_KEY, baseURL: baseURL || 'https://api.fireworks.ai/inference/v1' })(modelNameString),

3. Optionally, adjust the default structured output mode in the getDefaultMode function:

   1
   if (providerId === 'fireworks') {
   2
     return 'json'
   3
   }

4. Optionally, add a new logo under public/thirdparty/logos [blocked]

Contributing

As an open-source project, we welcome contributions from the community. If you are experiencing any bugs or want to add some improvements, please feel free to open an issue or pull request.