### Install Dependencies with uv

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Installs project dependencies using 'uv'. Supports installation from a 'pyproject.toml' file or a 'requirements.txt' file.

```bash
uv pip install .
```

```bash
uv pip install -r requirements.txt
```

--------------------------------

### Install Dependencies with pip

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Installs project dependencies using 'pip' after activating the virtual environment. Requires a 'requirements.txt' file.

```bash
pip install -r requirements.txt
```

--------------------------------

### Launch Project Server

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Starts the LemonAI project server. Ensure the virtual environment is activated and you are in the correct directory before running.

```python
# activate virtual environment
python src/server.py
```

--------------------------------

### Install Browser Plugin

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Installs the 'patchright' browser plugin for Chromium. This command should be run after activating the virtual environment.

```bash
# before this step, you need to activate virtual environment
patchright install chromium --with-deps --no-shell
```

--------------------------------

### Navigate to Project Directory

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Changes the current directory to the specified project path. This is a prerequisite for most installation and launch commands.

```bash
cd /path/to/browserUse
```

```bash
cd /path/to/browser_server
```

--------------------------------

### Create Virtual Environment with uv

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Creates a new Python virtual environment using the 'uv' package manager. This helps isolate project dependencies.

```bash
uv venv
```

--------------------------------

### Activate Virtual Environment

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/readme.md

Activates the created virtual environment. This ensures that subsequent commands use the environment's Python interpreter and packages.

```bash
source .venv/bin/activate
```

```bash
.venv\Scripts\activate
```

--------------------------------

### Install Chromium for Playwright

Source: https://github.com/hexdocom/lemonai/blob/main/resources/browser/install.md

Installs the Chromium browser required by Playwright for browser automation. This command should be run in the project's root directory before packaging.

```bash
npx playwright install chromium
```

--------------------------------

### Docker Deployment and Configuration (Bash)

Source: https://context7.com/hexdocom/lemonai/llms.txt

Provides commands to deploy Lemon AI using Docker. Includes pulling the runtime sandbox image, running the container with specific environment variables and volume mounts, and starting the application. It also shows how to use docker-compose for deployment.

```bash
# Pull the runtime sandbox image
docker pull hexdolemonai/lemon-runtime-sandbox:latest

# Run Lemon AI with Docker
docker run -it --rm --pull=always \
  --name lemon-app \
  --env DOCKER_HOST_ADDR=host.docker.internal \
  --env ACTUAL_HOST_WORKSPACE_PATH=${WORKSPACE_BASE:-$PWD/workspace} \
  --publish 5005:5005 \
  --add-host host.docker.internal:host-gateway \
  --volume /var/run/docker.sock:/var/run/docker.sock \
  --volume ~/.cache:/.cache \
  --volume ${WORKSPACE_BASE:-$PWD/workspace}:/workspace \
  --volume ${WORKSPACE_BASE:-$PWD/data}:/app/data \
  --interactive \
  --tty \
  hexdolemonai/lemon:latest make run
```

```yaml
services:
  lemon:
    image: hexdolemonai/lemon:latest
    container_name: lemon-app
    environment:
      - DOCKER_HOST_ADDR=host.docker.internal
      - ACTUAL_HOST_WORKSPACE_PATH=${WORKSPACE_BASE:-$PWD/workspace}
    ports:
      - "5005:5005"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
      - ~/.cache:/.cache
      - ${WORKSPACE_BASE:-$PWD/workspace}:/app/workspace
      - ${WORKSPACE_BASE:-$PWD/data}:/app/data
    stdin_open: true
    tty: true
    command: make run

# Start with docker-compose
docker-compose up -d
```

--------------------------------

### Common Action Sequences for AI Agent

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/browser_use/agent/prompt/base_system_prompt.md

Provides examples of common action sequences, such as form filling and navigation/extraction. These sequences illustrate how to chain actions for specific tasks.

```json
[
  {"input_text": {"index": 1, "text": "username"}},
  {"input_text": {"index": 2, "text": "password"}},
  {"click_element": {"index": 3}}
]
```

```json
[
  {"go_to_url": {"url": "https://example.com"}},
  {"extract_content": {"goal": "extract the names"}}
]
```

--------------------------------

### Run Lemon AI Editor with Docker

Source: https://github.com/hexdocom/lemonai/blob/main/README.md

This command pulls the latest Lemon AI runtime sandbox image and runs it as a container. It configures necessary environment variables, port mappings, and volume mounts for persistent storage and Docker socket integration. The container runs the 'make run' command to start the Lemon AI application.

```bash
docker pull hexdolemonai/lemon-runtime-sandbox:latest

docker run -it --rm --pull=always \
  --name lemon-app \
  --env DOCKER_HOST_ADDR=host.docker.internal \
  --env ACTUAL_HOST_WORKSPACE_PATH=${WORKSPACE_BASE:-$PWD/workspace} \
  --publish 5005:5005 \
  --add-host host.docker.internal:host-gateway \
  --volume /var/run/docker.sock:/var/run/docker.sock \
  --volume ~/.cache:/.cache \
  --volume ${WORKSPACE_BASE:-$PWD/workspace}:/workspace \
  --volume ${WORKSPACE_BASE:-$PWD/data}:/app/data \
  --interactive \
  --tty \
  hexdolemonai/lemon:latest make run
```

--------------------------------

### Get All Agents - Agent Management API - API

Source: https://context7.com/hexdocom/lemonai/llms.txt

Retrieves a list of all available AI agents. Requires an authorization token.

```bash
curl -X GET http://localhost:5005/api/agent \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Start Chat Conversation with cURL

Source: https://context7.com/hexdocom/lemonai/llms.txt

This snippet demonstrates initiating a chat conversation using cURL. It's designed for Q&A and general conversation, supporting conversation threading and context maintenance. Requires a bearer token.

```bash
curl -X POST http://localhost:5005/api/agent/chat \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "question": "What are the benefits of using Docker for development?",
    "conversation_id": "550e8400-e29b-41d4-a716-446655440000",
    "pid": -1,
    "model_id": 48
  }'
```

--------------------------------

### Get VSCode URL for Conversation Workspace (API)

Source: https://context7.com/hexdocom/lemonai/llms.txt

Retrieves the VSCode URL for a specific conversation workspace using a GET request to the runtime API. Requires a conversation ID and an authorization token. Returns a JSON object containing the workspace URL.

```bash
curl -X GET "http://localhost:5005/api/runtime/vscode-url?conversation_id=550e8400-e29b-41d4-a716-446655440000" \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Configure LLM Platforms and Models (POST, GET, PUT)

Source: https://context7.com/hexdocom/lemonai/llms.txt

API for managing LLM platforms (e.g., OpenAI, Anthropic) and their associated models. Supports creating, retrieving, updating platforms, and checking API availability. Also allows creating and listing models under specific platforms.

```bash
# Create a new platform
curl -X POST http://localhost:5005/api/platform \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "OpenAI",
    "logo_url": "https://example.com/openai.png",
    "source_type": "api"
  }'

# Get all platforms
curl -X GET http://localhost:5005/api/platform \
  -H "Authorization: Bearer <token>"

# Update platform with API credentials
curl -X PUT http://localhost:5005/api/platform/1 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "OpenAI",
    "api_key": "sk-...",
    "api_url": "https://api.openai.com/v1",
    "is_enabled": true
  }'

# Check API availability
curl -X POST http://localhost:5005/api/platform/check_api_availability \
  -H "Content-Type: application/json" \
  -d '{
    "base_url": "https://api.openai.com/v1",
    "api_key": "sk-...",
    "model": "gpt-4"
  }'

# Create a model under a platform
curl -X POST http://localhost:5005/api/model \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "platform_id": 1,
    "model_id": "gpt-4-turbo",
    "model_name": "GPT-4 Turbo",
    "group_name": "GPT-4",
    "model_types": ["chat", "code"]
  }'

# Get enabled models across all platforms
curl -X GET http://localhost:5005/api/model/enabled \
  -H "Authorization: Bearer <token>"

# Get models for a specific platform
curl -X GET http://localhost:5005/api/model/list/1 \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Manage Knowledge Base Entries (POST, GET, PUT, DELETE)

Source: https://context7.com/hexdocom/lemonai/llms.txt

API for managing agent knowledge entries, enabling self-evolving learning. Supports creating, retrieving (all for agent, specific entry), updating, and deleting knowledge. Requires authorization and agent ID for some operations.

```bash
# Create a knowledge entry
curl -X POST http://localhost:5005/api/knowledge \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "content": "When creating charts, always use responsive designs",
    "category": "best_practices",
    "agent_id": 1
  }'

# Get all knowledge for an agent
curl -X GET "http://localhost:5005/api/knowledge?agent_id=1" \
  -H "Authorization: Bearer <token>"

# Response:
# {
#   "code": 0,
#   "data": {
#     "personal": [{"id": 1, "content": "Custom knowledge..."}],
#     "system": [{"id": 2, "content": "Learned from conversations...", "is_learned": true}]
#   }
# }

# Get a specific knowledge entry
curl -X GET http://localhost:5005/api/knowledge/detail/1 \
  -H "Authorization: Bearer <token>"

# Update a knowledge entry
curl -X PUT http://localhost:5005/api/knowledge/1 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "content": "Updated best practice for chart creation",
    "category": "visualization"
  }'

# Delete a knowledge entry
curl -X DELETE http://localhost:5005/api/knowledge/1 \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Manage MCP Servers (POST, GET, PUT, DELETE)

Source: https://context7.com/hexdocom/lemonai/llms.txt

Endpoints for managing Model Context Protocol (MCP) servers, which extend agent capabilities. Supports creating, retrieving (all, active), updating, and deleting MCP server connections. Requires authorization.

```bash
# Create an MCP server connection
curl -X POST http://localhost:5005/api/mcp_server \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "GitHub Integration",
    "url": "https://mcp.github.com",
    "description": "GitHub repository operations",
    "activate": true,
    "type": "stdio",
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-github"],
    "env": {"GITHUB_TOKEN": "ghp_..."}
  }'

# Get all MCP servers
curl -X GET http://localhost:5005/api/mcp_server \
  -H "Authorization: Bearer <token>"

# Get active MCP servers only
curl -X GET http://localhost:5005/api/mcp_server/active \
  -H "Authorization: Bearer <token>"

# Update an MCP server
curl -X PUT http://localhost:5005/api/mcp_server/1 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "GitHub Integration v2",
    "activate": true
  }'

# Delete an MCP server
curl -X DELETE http://localhost:5005/api/mcp_server/1 \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Get VSCode URL for a conversation workspace

Source: https://context7.com/hexdocom/lemonai/llms.txt

Retrieves the VSCode URL for a given conversation ID, allowing direct access to the workspace in a VSCode environment.

```APIDOC
## GET /api/runtime/vscode-url

### Description
Get the VSCode URL for a conversation workspace.

### Method
GET

### Endpoint
/api/runtime/vscode-url

### Parameters
#### Query Parameters
- **conversation_id** (string) - Required - The ID of the conversation workspace.

#### Headers
- **Authorization** (string) - Required - Bearer token for authentication.

### Response
#### Success Response (200)
- **code** (integer) - Response code, 0 indicates success.
- **data** (object) - Contains the URL for the VSCode workspace.
  - **url** (string) - The URL to access the VSCode workspace.
```

--------------------------------

### Copy Chromium Browser Binaries to Resources (Mac)

Source: https://github.com/hexdocom/lemonai/blob/main/resources/browser/install.md

Copies the installed Chromium browser binaries from the Playwright cache to the project's resources directory. This is a necessary step for the search functionality to work on Mac systems.

```bash
cp -r ~/Library/Caches/ms-playwright/chromium_headless_shell-*/* resources/browser/chromium/
```

--------------------------------

### Docker Deployment

Source: https://context7.com/hexdocom/lemonai/llms.txt

Instructions for deploying Lemon AI using Docker, including pulling the image, running the container, and using docker-compose.

```APIDOC
## Docker Deployment

Deploy Lemon AI using Docker for quick setup with a sandboxed execution environment.

### Pulling the Image
```bash
docker pull hexdolemonai/lemon-runtime-sandbox:latest
```

### Running Lemon AI with Docker
```bash
docker run -it --rm --pull=always \
  --name lemon-app \
  --env DOCKER_HOST_ADDR=host.docker.internal \
  --env ACTUAL_HOST_WORKSPACE_PATH=${WORKSPACE_BASE:-$PWD/workspace} \
  --publish 5005:5005 \
  --add-host host.docker.internal:host-gateway \
  --volume /var/run/docker.sock:/var/run/docker.sock \
  --volume ~/.cache:/.cache \
  --volume ${WORKSPACE_BASE:-$PWD/workspace}:/workspace \
  --volume ${WORKSPACE_BASE:-$PWD/data}:/app/data \
  --interactive \
  --tty \
  hexdolemonai/lemon:latest make run
```

### Using Docker Compose

#### `docker-compose.yml`
```yaml
services:
  lemon:
    image: hexdolemonai/lemon:latest
    container_name: lemon-app
    environment:
      - DOCKER_HOST_ADDR=host.docker.internal
      - ACTUAL_HOST_WORKSPACE_PATH=${WORKSPACE_BASE:-$PWD/workspace}
    ports:
      - "5005:5005"
    extra_hosts:
      - "host.docker.internal:host-gateway"
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
      - ~/.cache:/.cache
      - ${WORKSPACE_BASE:-$PWD/workspace}:/app/workspace
      - ${WORKSPACE_BASE:-$PWD/data}:/app/data
    stdin_open: true
    tty: true
    command: make run
```

#### Starting with Docker Compose
```bash
docker-compose up -d
```
```

--------------------------------

### HTML Structure for Application Initialization Progress

Source: https://github.com/hexdocom/lemonai/blob/main/setup.html

This HTML snippet defines the structure for displaying the initialization steps and progress of the LemonAI application. It includes elements for step indicators, status messages, and a progress bar. The styling is handled by CSS.

```html
<div class="container">
  <div class="steps-container">
    <div class="step-item">
      <div class="step-number active">1</div>
      <div class="step-title">Check if Docker is Installed</div>
    </div>
    <div class="step-item">
      <div class="step-number">2</div>
      <div class="step-title">Check if Docker is Start</div>
    </div>
    <div class="step-item">
      <div class="step-number">3</div>
      <div class="step-title">Check if image exists</div>
    </div>
  </div>
  <div id="status-message">Initializing...</div>
  <div class="progress-bar-container">
    <div class="progress-bar" style="width: 0%;">0%</div>
  </div>
  <div id="error-message"></div>
</div>
```

--------------------------------

### CSS Styling for Application Initialization UI

Source: https://github.com/hexdocom/lemonai/blob/main/setup.html

This CSS code provides styling for the LemonAI application initialization interface. It defines the layout, appearance of steps, progress bar, and status messages to create a user-friendly progress indicator.

```css
body {
  font-family: Arial, sans-serif;
  text-align: center;
  padding-top: 50px;
  background-color: #f9f9f9;
}

.container {
  max-width: 600px;
  margin: 0 auto;
}

.steps-container {
  display: flex;
  justify-content: space-between;
  align-items: flex-start;
}

.step-item {
  width: calc(33.33% - 20px);
  position: relative;
  display: flex;
  flex-direction: column;
  align-content: center;
  align-items: center;
}

#progress-bar-container {
  display: none;
  width: 80%;
  margin: 20px auto;
  background-color: #f3f3f3;
  border-radius: 5px;
}

#progress-bar {
  width: 0%;
  height: 25px;
  background-color: #4CAF50;
  border-radius: 5px;
  text-align: center;
  line-height: 25px;
  color: white;
}

.step-number {
  font-size: 1em;
  font-weight: bold;
  color: #fff;
  width: 32px;
  height: 32px;
  margin-top: 10px;
  margin-bottom: 10px;
  border-radius: 100px;
  background: #d9d9d9;
  display: flex;
  justify-content: center;
  align-items: center;
}

#status-message {
  margin-top: 30px;
}

.step-number.active {
  background: #1677ff;
}

/* .step-number.success { color: #52c41a; } .step-number.failed { color: #f5222d; } */

.step-title {
  font-size: 16px;
  margin-bottom: 5px;
  color: #333;
  text-align: center;
}

.status-message {
  color: #777;
  text-align: center;
}

.progress-bar-container {
  width: 100%;
  background-color: #e0e0e0;
  border-radius: 5px;
  overflow: hidden;
  display: none;
  margin-top: 10px;
}

.progress-bar {
  height: 25px;
  background-color: #1890ff;
  border-radius: 5px;
  text-align: center;
  line-height: 25px;
  color: white;
  transition: width 1s ease-in-out;
}

.progress-bar.failed {
  background-color: #f5222d;
}

.arrow {
  position: absolute;
  right: 0;
  top: 50%;
  transform: translateY(-50%);
  font-size: 2em;
  color: #aaa;
}

.last-step .arrow {
  display: none;
}

#error-message {
  color: red;
  margin-top: 20px;
  display: none;
  font-size: 1.1em;
}
```

--------------------------------

### Get All Conversations with cURL

Source: https://context7.com/hexdocom/lemonai/llms.txt

This snippet demonstrates how to retrieve all conversations associated with a specific agent using cURL. It requires the agent ID as a query parameter and a bearer token for authorization.

```bash
curl -X GET "http://localhost:5005/api/conversation?agent_id=1" \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Manage Agents (GET, PUT, DELETE)

Source: https://context7.com/hexdocom/lemonai/llms.txt

Endpoints for retrieving, updating, and deleting specific agents. Requires an authorization token. The update endpoint accepts JSON payload for agent modifications.

```bash
curl -X GET http://localhost:5005/api/agent/1 \
  -H "Authorization: Bearer <token>"

curl -X PUT http://localhost:5005/api/agent/1 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "Advanced Data Agent",
    "describe": "Enhanced data analysis capabilities",
    "mcp_server_ids": [1, 2, 3]
  }'

curl -X DELETE http://localhost:5005/api/agent/1 \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Get Specific Conversation with cURL

Source: https://context7.com/hexdocom/lemonai/llms.txt

This snippet shows how to retrieve details for a specific conversation using its ID via cURL. It requires the conversation ID in the URL path and a bearer token for authentication.

```bash
curl -X GET http://localhost:5005/api/conversation/550e8400-e29b-41d4-a716-446655440000 \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### Environment Configuration

Source: https://context7.com/hexdocom/lemonai/llms.txt

Details on configuring Lemon AI's behavior using environment variables, covering storage, workspace, runtime, knowledge base, and platform API keys.

```APIDOC
## Environment Configuration

Configure Lemon AI behavior through environment variables.

### `.env` Configuration File Examples

#### Core Settings
- **STORAGE_PATH** (string) - SQLite database location (e.g., `data/database.sqlite`).
- **WORKSPACE_DIR** (string) - Workspace directory for conversations (e.g., `workspace`).
- **RUNTIME_TYPE** (string) - Specifies the runtime environment: `local`, `docker`, or `local-docker`.
- **ENABLE_KNOWLEDGE** (string) - Enables or disables the self-learning knowledge base (e.g., `ON` or `OFF`).

#### Platform API Configuration (Optional)
- **OPENAI_API_KEY** (string) - Your OpenAI API key.
- **ANTHROPIC_API_KEY** (string) - Your Anthropic API key.

#### Static File Serving (for Docker runtime)
- **STATIC_DOMAIN** (string) - Domain for serving static files (e.g., `static.lemonai.ai`).
- **STATIC_PROTOCOL** (string) - Protocol for static file serving (e.g., `https`).

### Example `.env` file:
```dotenv
STORAGE_PATH=data/database.sqlite
WORKSPACE_DIR=workspace
RUNTIME_TYPE=local-docker
ENABLE_KNOWLEDGE=ON

# Platform API configuration (optional)
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...

# Static file serving (for docker runtime)
STATIC_DOMAIN=static.lemonai.ai
STATIC_PROTOCOL=https
```
```

--------------------------------

### Get File Version History - File Editor API - API

Source: https://context7.com/hexdocom/lemonai/llms.txt

Retrieves the version history for a specific file within a conversation. Requires the conversation ID and file path in JSON format. Requires an authorization token.

```bash
curl -X POST http://localhost:5005/api/file/editor/versions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "conversation_id": "550e8400-e29b-41d4-a716-446655440000",
    "filepath": "Conversation_550e84/dashboard.html"
  }'
```

--------------------------------

### Platform and Model Configuration API

Source: https://context7.com/hexdocom/lemonai/llms.txt

Configure LLM platforms and manage models, including API key management and availability checks.

```APIDOC
## POST /api/platform

### Description
Creates a new LLM platform configuration.

### Method
POST

### Endpoint
/api/platform

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **name** (string) - Required - The name of the platform (e.g., "OpenAI").
- **logo_url** (string) - Optional - URL for the platform's logo.
- **source_type** (string) - Required - The type of source (e.g., "api", "local").

### Request Example
```bash
curl -X POST http://localhost:5005/api/platform \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{ "name": "OpenAI", "logo_url": "https://example.com/openai.png", "source_type": "api" }'
```

### Response
#### Success Response (200)
- **id** (integer) - The unique identifier of the created platform.
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "id": 1,
  "message": "Platform created successfully"
}
```

## GET /api/platform

### Description
Retrieves a list of all configured LLM platforms.

### Method
GET

### Endpoint
/api/platform

### Parameters
None

### Request Example
```bash
curl -X GET http://localhost:5005/api/platform \
  -H "Authorization: Bearer <token>"
```

### Response
#### Success Response (200)
- **platforms** (array of objects) - A list of platform configurations.

#### Response Example
```json
[
  {
    "id": 1,
    "name": "OpenAI",
    "logo_url": "https://example.com/openai.png",
    "source_type": "api"
  }
]
```

## PUT /api/platform/{id}

### Description
Updates an existing platform configuration, potentially adding API credentials.

### Method
PUT

### Endpoint
/api/platform/{id}

### Parameters
#### Path Parameters
- **id** (integer) - Required - The unique identifier of the platform to update.

#### Query Parameters
None

#### Request Body
- **name** (string) - Optional - The new name for the platform.
- **api_key** (string) - Optional - The API key for the platform.
- **api_url** (string) - Optional - The API endpoint URL.
- **is_enabled** (boolean) - Optional - Whether the platform is enabled.

### Request Example
```bash
curl -X PUT http://localhost:5005/api/platform/1 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{ "name": "OpenAI", "api_key": "sk-...", "api_url": "https://api.openai.com/v1", "is_enabled": true }'
```

### Response
#### Success Response (200)
- **message** (string) - Confirmation message of the update.

#### Response Example
```json
{
  "message": "Platform updated successfully"
}
```

## POST /api/platform/check_api_availability

### Description
Checks the availability of an LLM API endpoint with provided credentials and model.

### Method
POST

### Endpoint
/api/platform/check_api_availability

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **base_url** (string) - Required - The base URL of the API.
- **api_key** (string) - Required - The API key for authentication.
- **model** (string) - Required - The model to check availability for.

### Request Example
```bash
curl -X POST http://localhost:5005/api/platform/check_api_availability \
  -H "Content-Type: application/json" \
  -d '{ "base_url": "https://api.openai.com/v1", "api_key": "sk-...", "model": "gpt-4" }'
```

### Response
#### Success Response (200)
- **available** (boolean) - Indicates if the API is available.
- **message** (string) - Status message.

#### Response Example
```json
{
  "available": true,
  "message": "API is available"
}
```

## POST /api/model

### Description
Creates a new model configuration under a specific platform.

### Method
POST

### Endpoint
/api/model

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **platform_id** (integer) - Required - The ID of the platform this model belongs to.
- **model_id** (string) - Required - The unique identifier for the model.
- **model_name** (string) - Required - The human-readable name of the model.
- **group_name** (string) - Optional - The group the model belongs to.
- **model_types** (array of strings) - Optional - Types of the model (e.g., "chat", "code").

### Request Example
```bash
curl -X POST http://localhost:5005/api/model \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{ "platform_id": 1, "model_id": "gpt-4-turbo", "model_name": "GPT-4 Turbo", "group_name": "GPT-4", "model_types": ["chat", "code"] }'
```

### Response
#### Success Response (200)
- **id** (integer) - The unique identifier of the created model.
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "id": 1,
  "message": "Model created successfully"
}
```

## GET /api/model/enabled

### Description
Retrieves a list of all enabled models across all platforms.

### Method
GET

### Endpoint
/api/model/enabled

### Parameters
None

### Request Example
```bash
curl -X GET http://localhost:5005/api/model/enabled \
  -H "Authorization: Bearer <token>"
```

### Response
#### Success Response (200)
- **models** (array of objects) - A list of enabled model configurations.

#### Response Example
```json
[
  {
    "id": 1,
    "platform_id": 1,
    "model_id": "gpt-4",
    "model_name": "GPT-4"
  }
]
```

## GET /api/model/list/{platform_id}

### Description
Retrieves a list of models available for a specific platform.

### Method
GET

### Endpoint
/api/model/list/{platform_id}

### Parameters
#### Path Parameters
- **platform_id** (integer) - Required - The ID of the platform whose models to retrieve.

#### Query Parameters
None

#### Request Body
None

### Request Example
```bash
curl -X GET http://localhost:5005/api/model/list/1 \
  -H "Authorization: Bearer <token>"
```

### Response
#### Success Response (200)
- **models** (array of objects) - A list of model configurations for the specified platform.

#### Response Example
```json
[
  {
    "id": 1,
    "model_id": "gpt-4",
    "model_name": "GPT-4"
  }
]
```
```

--------------------------------

### Create a New Agent - Agent Management API - API

Source: https://context7.com/hexdocom/lemonai/llms.txt

Creates a new AI agent with specified name, description, MCP server IDs, and public status. Requires an authorization token.

```bash
curl -X POST http://localhost:5005/api/agent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{
    "name": "Data Analysis Agent",
    "describe": "Specialized in data visualization and analysis",
    "mcp_server_ids": [1, 2],
    "is_public": true
  }'
```

--------------------------------

### Task Completion and 'Done' Action

Source: https://github.com/hexdocom/lemonai/blob/main/browser_server/browser_use/agent/prompt/base_system_prompt.md

Explains the usage of the 'done' action for task completion, including conditions for its use and how to include gathered information. It also covers handling partial completion and success status.

```json
// Example of 'done' action when task is fully completed
{
  "action": [
    { "done": { "success": true, "text": "All requested information gathered." } }
  ]
}

// Example of 'done' action when task is partially completed (e.g., reached max steps)
{
  "action": [
    { "done": { "success": false, "text": "Reached max steps. Gathered partial information: ..." } }
  ]
}
```

--------------------------------

### List All Uploaded Files - API

Source: https://context7.com/hexdocom/lemonai/llms.txt

Retrieves a list of all files that have been uploaded. Requires an authorization token.

```bash
curl -X GET http://localhost:5005/api/file/list \
  -H "Authorization: Bearer <token>"
```

--------------------------------

### File Editor and Version Control API

Source: https://context7.com/hexdocom/lemonai/llms.txt

APIs for editing files, managing versions, and reading file content.

```APIDOC
## PUT /api/file/editor

### Description
Updates the content of a file, creating a new version.

### Method
PUT

### Endpoint
`/api/file/editor`

### Parameters
#### Request Body
- **path** (string) - Required - The path to the file.
- **content** (string) - Required - The new content for the file.
- **conversation_id** (string) - Required - The ID of the conversation the file belongs to.

### Request Example
```json
{
  "path": "Conversation_550e84/dashboard.html",
  "content": "<html><body>Updated content</body></html>",
  "conversation_id": "550e8400-e29b-41d4-a716-446655440000"
}
```

### Response
#### Success Response (200)
- **data** (object) - Information about the updated file version.
- **data.id** (integer) - The ID of the file.
- **data.version** (integer) - The new version number of the file.
- **data.filepath** (string) - The path of the file.

#### Response Example
```json
{
  "code": 0,
  "data": {"id": 5, "version": 3, "filepath": "Conversation_550e84/dashboard.html"}
}
```

## POST /api/file/editor/read

### Description
Reads the content of a file.

### Method
POST

### Endpoint
`/api/file/editor/read`

### Parameters
#### Request Body
- **path** (string) - Required - The path to the file.

### Request Example
```json
{
  "path": "Conversation_550e84/dashboard.html"
}
```

## POST /api/file/editor/versions

### Description
Retrieves the version history for a specific file.

### Method
POST

### Endpoint
`/api/file/editor/versions`

### Parameters
#### Request Body
- **conversation_id** (string) - Required - The ID of the conversation the file belongs to.
- **filepath** (string) - Required - The path of the file.

### Request Example
```json
{
  "conversation_id": "550e8400-e29b-41d4-a716-446655440000",
  "filepath": "Conversation_550e84/dashboard.html"
}
```

## POST /api/file/editor/switch-version

### Description
Switches a file to a specific version.

### Method
POST

### Endpoint
`/api/file/editor/switch-version`

### Parameters
#### Request Body
- **version_id** (integer) - Required - The ID of the version to switch to.
- **conversation_id** (string) - Required - The ID of the conversation the file belongs to.
- **filepath** (string) - Required - The path of the file.

### Request Example
```json
{
  "version_id": 2,
  "conversation_id": "550e8400-e29b-41d4-a716-446655440000",
  "filepath": "Conversation_550e84/dashboard.html"
}
```
```

--------------------------------

### MCP Server Integration API

Source: https://context7.com/hexdocom/lemonai/llms.txt

Manage Model Context Protocol (MCP) servers for extending agent capabilities.

```APIDOC
## POST /api/mcp_server

### Description
Creates a new MCP server connection.

### Method
POST

### Endpoint
/api/mcp_server

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **name** (string) - Required - The name of the MCP server.
- **url** (string) - Optional - The URL of the MCP server.
- **description** (string) - Optional - A description of the server.
- **activate** (boolean) - Optional - Whether to activate the server upon creation.
- **type** (string) - Optional - The type of server (e.g., "stdio").
- **command** (string) - Optional - The command to run the server.
- **args** (array of strings) - Optional - Arguments for the command.
- **env** (object) - Optional - Environment variables for the server process.

### Request Example
```bash
curl -X POST http://localhost:5005/api/mcp_server \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{ "name": "GitHub Integration", "url": "https://mcp.github.com", "description": "GitHub repository operations", "activate": true, "type": "stdio", "command": "npx", "args": ["-y", "@modelcontextprotocol/server-github"], "env": {"GITHUB_TOKEN": "ghp_..."} }'
```

### Response
#### Success Response (200)
- **id** (integer) - The unique identifier of the created MCP server.
- **message** (string) - Confirmation message.

#### Response Example
```json
{
  "id": 1,
  "message": "MCP server created successfully"
}
```

## GET /api/mcp_server

### Description
Retrieves a list of all configured MCP servers.

### Method
GET

### Endpoint
/api/mcp_server

### Parameters
None

### Request Example
```bash
curl -X GET http://localhost:5005/api/mcp_server \
  -H "Authorization: Bearer <token>"
```

### Response
#### Success Response (200)
- **servers** (array of objects) - A list of MCP server configurations.

#### Response Example
```json
[
  {
    "id": 1,
    "name": "GitHub Integration",
    "url": "https://mcp.github.com"
  }
]
```

## GET /api/mcp_server/active

### Description
Retrieves a list of only the active MCP servers.

### Method
GET

### Endpoint
/api/mcp_server/active

### Parameters
None

### Request Example
```bash
curl -X GET http://localhost:5005/api/mcp_server/active \
  -H "Authorization: Bearer <token>"
```

### Response
#### Success Response (200)
- **servers** (array of objects) - A list of active MCP server configurations.

#### Response Example
```json
[
  {
    "id": 1,
    "name": "GitHub Integration",
    "url": "https://mcp.github.com"
  }
]
```

## PUT /api/mcp_server/{id}

### Description
Updates an existing MCP server configuration.

### Method
PUT

### Endpoint
/api/mcp_server/{id}

### Parameters
#### Path Parameters
- **id** (integer) - Required - The unique identifier of the MCP server to update.

#### Query Parameters
None

#### Request Body
- **name** (string) - Optional - The new name for the MCP server.
- **activate** (boolean) - Optional - Whether to activate or deactivate the server.

### Request Example
```bash
curl -X PUT http://localhost:5005/api/mcp_server/1 \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <token>" \
  -d '{ "name": "GitHub Integration v2", "activate": true }'
```

### Response
#### Success Response (200)
- **message** (string) - Confirmation message of the update.

#### Response Example
```json
{
  "message": "MCP server updated successfully"
}
```

## DELETE /api/mcp_server/{id}

### Description
Deletes a specific MCP server configuration.

### Method
DELETE

### Endpoint
/api/mcp_server/{id}

### Parameters
#### Path Parameters
- **id** (integer) - Required - The unique identifier of the MCP server to delete.

#### Query Parameters
None

#### Request Body
None

### Request Example
```bash
curl -X DELETE http://localhost:5005/api/mcp_server/1 \
  -H "Authorization: Bearer <token>"
```

### Response
#### Success Response (200)
- **message** (string) - Confirmation message of the deletion.

#### Response Example
```json
{
  "message": "MCP server deleted successfully"
}
```
```

--------------------------------

### Environment Configuration (.env)

Source: https://context7.com/hexdocom/lemonai/llms.txt

Defines environment variables for configuring Lemon AI. Includes settings for storage path, workspace directory, runtime type, knowledge base enablement, platform API keys, and static file serving.

```dotenv
# .env configuration file
STORAGE_PATH=data/database.sqlite    # SQLite database location
WORKSPACE_DIR=workspace              # Workspace directory for conversations
RUNTIME_TYPE=local-docker            # Runtime: local, docker, local-docker
ENABLE_KNOWLEDGE=ON                  # Enable self-learning knowledge base

# Platform API configuration (optional)
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...

# Static file serving (for docker runtime)
STATIC_DOMAIN=static.lemonai.ai
STATIC_PROTOCOL=https
```