### Install Midday CLI

Source: https://github.com/midday-ai/midday/blob/main/packages/cli/README.md

Install the Midday CLI using npx or npm. Global installation is recommended for frequent use.

```bash
npx @midday-ai/cli@latest
```

```bash
npm install -g @midday-ai/cli
```

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

### Install @midday/categories

Source: https://github.com/midday-ai/midday/blob/main/packages/categories/README.md

Install the package using npm.

```bash
npm install @midday/categories
```

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

### Start API in Development Mode

Source: https://github.com/midday-ai/midday/blob/main/apps/api/README.md

Run this command to start the API server in development mode.

```bash
bun dev
```

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

### Start Live Timer

Source: https://context7.com/midday-ai/midday/llms.txt

Start a live timer for a specified project, optionally with a description. Requires `Authorization` and `Content-Type` headers.

```bash
# Start a live timer
curl -X POST https://api.midday.ai/tracker-entries/timer/start \
  -H "Authorization: Bearer mid_live_abc123" \
  -H "Content-Type: application/json" \
  -d '{ "projectId": "proj_uuid", "description": "Client call" }'
```

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

### Start API in Production Mode

Source: https://github.com/midday-ai/midday/blob/main/apps/api/README.md

Run this command to start the API server in production mode.

```bash
bun start
```

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

### NPM SDK Installation

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/api-reference.mdx

Install the Midday SDK using npm.

```APIDOC
## NPM SDK

### Description
Install the official Midday SDK for Node.js environments.

### Installation
```bash
npm install @midday-ai/sdk
```
```

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

### Authorization Endpoint Request Example

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/oauth-api-endpoints.mdx

Construct a GET request to the authorization endpoint to initiate the OAuth flow. Ensure all required parameters, including PKCE parameters if used, are correctly included.

```http
https://app.midday.ai/oauth/authorize?
  response_type=code&
  client_id=mid_client_abc123&
  redirect_uri=https://yourapp.com/callback&
  scope=transactions.read%20invoices.read&
  state=xyz789&
  code_challenge=E9Melhoa2OwvFrEMTJguCHaoeK1t8URWbuGJSstw-cM&
  code_challenge_method=S256
```

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

### Start Live Timer

Source: https://context7.com/midday-ai/midday/llms.txt

Starts a live timer for a project. This is useful for tracking time in real-time.

```APIDOC
## POST /tracker-entries/timer/start

### Description
Starts a live timer for a project. This is useful for tracking time in real-time.

### Method
POST

### Endpoint
/tracker-entries/timer/start

### Parameters
#### Request Body
- **projectId** (string) - Required - The ID of the project to start the timer for.
- **description** (string) - Optional - A description for the timer entry.

### Request Example
```bash
curl -X POST https://api.midday.ai/tracker-entries/timer/start \
  -H "Authorization: Bearer mid_live_abc123" \
  -H "Content-Type: application/json" \
  -d '{ "projectId": "proj_uuid", "description": "Client call" }'
```
```

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

### Install Midday NPM SDK

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/api-reference.mdx

Install the official Midday SDK for Node.js and browser environments using npm.

```bash
npm install @midday-ai/sdk
```

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

### Install Midday MCP Package

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/api-reference.mdx

Install the Midday MCP package for AI tool integrations using npx.

```bash
npx @midday-ai/mcp
```

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

### Start Redis with Docker

Source: https://github.com/midday-ai/midday/blob/main/apps/api/README.md

Use this command to start a Redis instance locally using Docker for development purposes.

```bash
docker run -d --name redis -p 6379:6379 redis:alpine
```

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

### MCP Package Installation

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/api-reference.mdx

Install the MCP package for AI tool integrations.

```APIDOC
## MCP Package

### Description
Install the MCP package for integrating with AI tools.

### Installation
```bash
npx @midday-ai/mcp
```
```

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

### Provider Balance Normalization Example

Source: https://github.com/midday-ai/midday/blob/main/docs/runway-burn-rate-analysis.md

Illustrates how raw API responses for balances are stored in the database, highlighting normalization for GoCardless.

```plaintext
Provider | Raw API Response | Stored In Database
| Plaid | Positive (`1000` = $1000 owed) | Positive
| GoCardless | Negative (`-1500` = $1500 owed) | **Normalized to positive**
| EnableBanking | Positive (ISO 20022 `current`) | Positive (normalized as safety)
| Teller | Positive | Positive (normalized as safety)
```

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

### Run Midday Desktop App in Development Mode

Source: https://github.com/midday-ai/midday/blob/main/apps/desktop/README.md

Execute this command to run the app in development mode, which loads the local development server at http://localhost:3001. Ensure you have Bun installed.

```bash
bun run tauri:dev
```

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

### Get Platform-Specific Instructions

Source: https://github.com/midday-ai/midday/blob/main/docs/bot-messaging-integration.md

Retrieves platform-specific instructions, appending formatting hints to the system prompt for enhanced user experience.

```typescript
import { getPlatformInstructions, BotPlatform } from "@midday/bot";

const platform: BotPlatform = "telegram"; // or "whatsapp", "slack"
const instructions = getPlatformInstructions(platform);
console.log(instructions);
```

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

### Initialize Midday SDK with Access Token

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/build-oauth-app.mdx

Instantiate the Midday SDK using the obtained access token to make authenticated API requests. This example shows how to list transactions, invoices, and retrieve financial metrics.

```typescript
import { Midday } from "@midday-ai/sdk";

const midday = new Midday({
  token: "mid_at_xxxxx", // Access token from OAuth flow
});

// List transactions
const transactions = await midday.transactions.list({
  pageSize: 50,
});

// Get invoices
const invoices = await midday.invoices.list({
  pageSize: 20,
});

// Get financial metrics
const revenue = await midday.metrics.revenue({
  from: "2024-01-01",
  to: "2024-12-31",
});
```

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

### Track Export Transactions Event with LogSnag

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/updates/posts/lognsag.mdx

An example of using the `setupLogSnag` function within a server action to track a 'transactions export' event, including user identification and event details.

```typescript
export const exportTransactionsAction = action(
  exportTransactionsSchema,
  async (transactionIds) => {
    const user = await getUser();

    const event = await client.sendEvent({
      name: Events.TRANSACTIONS_EXPORT,
      payload: {
        transactionIds,
        teamId: user.data.team_id,
        locale: user.data.locale,
      },
    });

    const logsnag = await setupLogSnag({
      userId: user.data.id,
      fullName: user.data.full_name,
    });

    logsnag.track({
      event: LogEvents.ExportTransactions.name,
      icon: LogEvents.ExportTransactions.icon,
      channel: LogEvents.ExportTransactions.channel,
    });

    return event;
  }
);

```

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

### Minimal Read Scope Example

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/oauth-scopes.mdx

Request only the necessary read scopes for specific functionalities, such as a spending tracker.

```text
transactions.read
```

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

### Generate Business Insights with InsightsService

Source: https://github.com/midday-ai/midday/blob/main/packages/insights/README.md

Use the `createInsightsService` function to initialize the service and then call `generateInsight` with the appropriate parameters to get business insights. Ensure the necessary imports from `@midday/insights` and `@midday/db/client` are included.

```typescript
import { createInsightsService } from "@midday/insights";
import { db } from "@midday/db/client";

const service = createInsightsService(db);

const result = await service.generateInsight({
  teamId: "team-uuid",
  periodType: "weekly",
  periodStart: new Date("2024-01-08"),
  periodEnd: new Date("2024-01-14"),
  periodLabel: "Week 2, 2024",
  periodYear: 2024,
  periodNumber: 2,
  currency: "USD",
});

// Result contains:
// - selectedMetrics: Top 4 most relevant metrics
// - allMetrics: Full metrics snapshot
// - anomalies: Detected issues/alerts
// - activity: Invoice, time tracking, customer activity
// - content: AI-generated narrative (sentiment, opener, story, actions)
```

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

### Full TypeScript OAuth Implementation

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/build-oauth-app.mdx

An example Express.js application demonstrating the full OAuth 2.0 flow with Midday.ai, including redirecting to authorization, handling callbacks, exchanging codes for tokens, and using the SDK.

```typescript
import express from "express";
import crypto from "crypto";
import { Midday } from "@midday-ai/sdk";

const app = express();

const CLIENT_ID = process.env.MIDDAY_CLIENT_ID!;
const CLIENT_SECRET = process.env.MIDDAY_CLIENT_SECRET!;
const REDIRECT_URI = "http://localhost:3000/callback";

// In production, use a proper session store
const sessions = new Map<string, { state: string; accessToken?: string }>();

// Step 1: Redirect to authorization
app.get("/connect", (req, res) => {
  const sessionId = crypto.randomUUID();
  const state = crypto.randomUUID();
  
  sessions.set(sessionId, { state });
  res.cookie("session_id", sessionId);
  
  const authUrl = new URL("https://app.midday.ai/oauth/authorize");
  authUrl.searchParams.set("response_type", "code");
  authUrl.searchParams.set("client_id", CLIENT_ID);
  authUrl.searchParams.set("redirect_uri", REDIRECT_URI);
  authUrl.searchParams.set("scope", "transactions.read invoices.read");
  authUrl.searchParams.set("state", state);
  
  res.redirect(authUrl.toString());
});

// Step 2: Handle callback and exchange code for tokens
app.get("/callback", async (req, res) => {
  const { code, state, error } = req.query;
  const sessionId = req.cookies.session_id;
  const session = sessions.get(sessionId);
  
  if (error) {
    return res.send("Authorization denied");
  }
  
  // Verify state matches
  if (state !== session?.state) {
    return res.status(400).send("Invalid state");
  }
  
  // Exchange code for tokens
  const response = await fetch("https://api.midday.ai/v1/oauth/token", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({
      grant_type: "authorization_code",
      code,
      redirect_uri: REDIRECT_URI,
      client_id: CLIENT_ID,
      client_secret: CLIENT_SECRET,
    }),
  });
  
  const tokens = await response.json();
  session.accessToken = tokens.access_token;
  
  res.send("Connected successfully!");
});

// Step 3: Use the SDK with the access token
app.get("/transactions", async (req, res) => {
  const sessionId = req.cookies.session_id;
  const session = sessions.get(sessionId);
  
  if (!session?.accessToken) {
    return res.status(401).send("Not connected");
  }
  
  const midday = new Midday({
    token: session.accessToken,
  });
  
  const result = await midday.transactions.list({
    pageSize: 50,
  });
  
  res.json(result);
});

app.listen(3000, () => {
  console.log("Server running on http://localhost:3000");
});

```

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

### Setup LogSnag with User Identification

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/updates/posts/lognsag.mdx

Initializes LogSnag, optionally identifying the user with their ID and name if consent is given. It also wraps the track method to conditionally include the user ID.

```typescript
export const setupLogSnag = async (options?: Props) => {
  const { userId, fullName } = options ?? {};
  const consent = cookies().get(Cookies.TrackingConsent)?.value === "0";

  const logsnag = new LogSnag({
    token: process.env.LOGSNAG_PRIVATE_TOKEN!,
    project: process.env.NEXT_PUBLIC_LOGSNAG_PROJECT!,
    disableTracking: Boolean(process.env.NEXT_PUBLIC_LOGSNAG_DISABLED!),
  });

  if (consent && userId && fullName) {
    await logsnag.identify({
      user_id: userId,
      properties: {
        name: fullName,
      },
    });
  }

  return {
    ...logsnag,
    track: (options: TrackOptions) =>
      logsnag.track({
        ...options,
        user_id: consent ? userId : undefined,
      }),
  };
};

```

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

### Build Midday Desktop App for Production

Source: https://github.com/midday-ai/midday/blob/main/apps/desktop/README.md

Use this command to create a production build of the Midday Desktop App. Requires Bun.

```bash
bun run tauri:build:prod
```

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

### Build Midday Desktop App for Staging

Source: https://github.com/midday-ai/midday/blob/main/apps/desktop/README.md

Use this command to create a staging build of the Midday Desktop App. Requires Bun.

```bash
bun run tauri:build:staging
```

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

### Build Midday Desktop App for Development

Source: https://github.com/midday-ai/midday/blob/main/apps/desktop/README.md

Use this command to create a development build of the Midday Desktop App. Requires Bun.

```bash
bun run tauri:build
```

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

### Run Midday Desktop App in Staging Mode

Source: https://github.com/midday-ai/midday/blob/main/apps/desktop/README.md

Execute this command to run the app in staging mode, loading the staging environment from https://beta.midday.ai. Requires Bun.

```bash
bun run tauri:staging
```

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

### Handle Assistant Thread Started Event

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/updates/posts/slack-assistant.mdx

This function is triggered when a new assistant thread starts. It sends a welcome message and suggests initial prompts to the user.

```typescript
import type { AssistantThreadStartedEvent, WebClient } from "@slack/web-api";

export async function assistantThreadStarted(
  event: AssistantThreadStartedEvent,
  client: WebClient,
) {
  const prompts = [
    { title: "What’s my runway?", message: "What's my runway?" },
    // Additional prompts...
  ];

  try {
    await client.chat.postMessage({
      channel: event.assistant_thread.channel_id,
      thread_ts: event.assistant_thread.thread_ts,
      text: "Welcome! I'm your financial assistant. Here are some suggestions:",
    });

    await client.assistant.threads.setSuggestedPrompts({
      channel_id: event.assistant_thread.channel_id,
      thread_ts: event.assistant_thread.thread_ts,
      prompts: prompts.sort(() => 0.5 - Math.random()).slice(0, 4),
    });
  } catch (error) {
    console.error("Error in assistant thread:", error);
    await client.assistant.threads.setStatus({
      channel_id: event.assistant_thread.channel_id,
      thread_ts: event.assistant_thread.thread_ts,
      status: "Something went wrong",
    });
  }
}
```

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

### Get Payment Health Score

Source: https://context7.com/midday-ai/midday/llms.txt

Retrieves the overall payment health score.

```APIDOC
## GET /invoices/payment-status

### Description
Gets the payment health score.

### Method
GET

### Endpoint
https://api.midday.ai/invoices/payment-status
```

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

### Get payment health score

Source: https://context7.com/midday-ai/midday/llms.txt

Retrieve the overall payment health score for your invoices.

```bash
curl "https://api.midday.ai/invoices/payment-status" \
  -H "Authorization: Bearer mid_live_abc123"
```

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

### Get Single Transaction

Source: https://context7.com/midday-ai/midday/llms.txt

Retrieve details for a specific transaction using its unique ID.

```bash
curl "https://api.midday.ai/transactions/txn_uuid" \
  -H "Authorization: Bearer mid_live_abc123"
```

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

### Update Invoice Status

Source: https://context7.com/midday-ai/midday/llms.txt

Updates the status of an existing invoice, for example, to mark it as paid.

```APIDOC
## PUT /invoices/{invoiceId}

### Description
Updates an invoice, including its status.

### Method
PUT

### Endpoint
https://api.midday.ai/invoices/{invoiceId}

### Parameters
#### Path Parameters
- **invoiceId** (string) - Required - The ID of the invoice to update.

### Request Body
- **updateData** (object) - Required - The data to update.
  - **status** (string) - Required - The new status for the invoice (e.g., `paid`).
  - **paidAt** (string) - Optional - The date and time the invoice was paid (ISO 8601 format), required if status is `paid`.

### Request Example
```json
{ "status": "paid", "paidAt": "2024-07-05T14:30:00Z" }
```
```

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

### Initialize Theme Preference

Source: https://github.com/midday-ai/midday/blob/main/packages/workbench/src/ui/index.html

Sets the initial theme (light or dark) based on local storage or system preference. This script runs immediately to prevent theme flash.

```javascript
(() => {
  var stored = localStorage.getItem('workbench:theme');
  var isDark = stored ? stored === 'dark' : window.matchMedia('(prefers-color-scheme: dark)').matches;
  if (isDark) document.documentElement.classList.add('dark');
})();
/* Prevent flash - set background immediately */
```

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

### Get Invoice Summary

Source: https://context7.com/midday-ai/midday/llms.txt

Retrieves a summary of invoices grouped by status, including counts and total amounts.

```APIDOC
## GET /invoices/summary

### Description
Gets a summary of invoices by status.

### Method
GET

### Endpoint
https://api.midday.ai/invoices/summary

### Parameters
#### Query Parameters
- **statuses[]** (string) - Optional - Filters the summary by one or more statuses (e.g., `unpaid`, `overdue`).

### Response
#### Success Response (200)
- **data** (array) - An array of summary objects, each containing status, count, totalAmount, and currency.
```

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

### Create a customer

Source: https://context7.com/midday-ai/midday/llms.txt

Add a new customer to the system. This data can be used to auto-populate invoice details.

```bash
curl -X POST https://api.midday.ai/customers \
  -H "Authorization: Bearer mid_live_abc123" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Acme Corp",
    "email": "billing@acme.com",
    "website": "https://acme.com",
    "address": "123 Main St",
    "city": "San Francisco",
    "state": "CA",
    "zip": "94105",
    "country": "US",
    "vatNumber": "US123456789"
  }'
```

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

### Get invoice summary by status

Source: https://context7.com/midday-ai/midday/llms.txt

Fetch a summary of invoices, including counts and total amounts, for specified statuses.

```bash
curl "https://api.midday.ai/invoices/summary?statuses[]=unpaid&statuses[]=overdue" \
  -H "Authorization: Bearer mid_live_abc123"
```

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

### Run Midday Desktop App in Production Mode

Source: https://github.com/midday-ai/midday/blob/main/apps/desktop/README.md

Execute this command to run the app in production mode, loading the production environment from https://app.midday.ai. Requires Bun.

```bash
bun run tauri:prod
```

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

### Using Midday SDK with OAuth Access Token

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/oauth-api-endpoints.mdx

Demonstrates how to initialize the Midday SDK with an OAuth access token and make various API requests such as listing transactions, invoices, and retrieving financial metrics.

```typescript
import { Midday } from "@midday-ai/sdk";

const midday = new Midday({
  token: accessToken, // OAuth access token
});

// List transactions
const transactions = await midday.transactions.list({
  pageSize: 50,
});

// Get invoices
const invoices = await midday.invoices.list({
  statuses: ["unpaid", "overdue"],
});

// Get financial metrics
const profit = await midday.metrics.profit({
  from: "2024-01-01",
  to: "2024-12-31",
});
```

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

### Group Related Scopes Example

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/oauth-scopes.mdx

Group related scopes together, such as requesting both invoice and customer scopes if your application deals with both.

```text
invoices.read invoices.write customers.read
```

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

### Configure Claude Desktop MCP Server

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/assistant-mcp.mdx

Add this configuration to your Claude Desktop's `claude_desktop_config.json` file to enable Midday integration. Ensure you replace `your-api-key-here` with your actual Midday API key.

```json
{
  "mcpServers": {
    "midday": {
      "command": "npx",
      "args": ["-y", "@midday-ai/mcp"],
      "env": {
        "MIDDAY_API_KEY": "your-api-key-here"
      }
    }
  }
}
```

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

### Get Pre-signed URL for Transaction Attachment

Source: https://context7.com/midday-ai/midday/llms.txt

Generates a pre-signed URL to download a transaction attachment. The URL is valid for 60 seconds.

```APIDOC
## POST /transactions/{transactionId}/attachments/{attachmentId}/presigned-url

### Description
Gets a pre-signed URL for a transaction attachment.

### Method
POST

### Endpoint
https://api.midday.ai/transactions/{transactionId}/attachments/{attachmentId}/presigned-url

### Parameters
#### Path Parameters
- **transactionId** (string) - Required - The ID of the transaction.
- **attachmentId** (string) - Required - The ID of the attachment.

#### Query Parameters
- **download** (boolean) - Optional - Set to `true` to indicate a download request.
```

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

### Get pre-signed URL for transaction attachment

Source: https://context7.com/midday-ai/midday/llms.txt

Obtain a temporary URL to download a transaction attachment. The URL is valid for 60 seconds.

```bash
curl -X POST "https://api.midday.ai/transactions/txn_uuid/attachments/att_uuid/presigned-url?download=true" \
  -H "Authorization: Bearer mid_live_abc123"
```

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

### Readiness check for service dependencies

Source: https://context7.com/midday-ai/midday/llms.txt

Check the readiness of the service, including its dependencies like database and Redis. Returns 'ok' or 'degraded' status.

```bash
curl https://api.midday.ai/health/ready
```

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

### Retrieve Category Colors

Source: https://github.com/midday-ai/midday/blob/main/packages/categories/README.md

Import and use functions to get the predefined color for any category or access the complete map of category colors.

```typescript
import { getCategoryColor, CATEGORY_COLOR_MAP } from '@midday/categories';

// Get color for any category
const revenueColor = getCategoryColor('revenue'); // "#00D084" (Green)
const officeSuppliesColor = getCategoryColor('office-supplies'); // "#8ED1FC" (Sky Blue)

// Access the complete color map
const allColors = CATEGORY_COLOR_MAP;
```

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

### Database Configuration

Source: https://github.com/midday-ai/midday/blob/main/apps/api/README.md

Configure these environment variables with your PostgreSQL database URLs for primary and replica instances across different regions.

```bash
DATABASE_PRIMARY_URL=postgresql://...
DATABASE_FRA_URL=postgresql://...  # EU replica
DATABASE_IAD_URL=postgresql://...  # US East replica
DATABASE_SJC_URL=postgresql://...  # US West replica
```

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

### Time Tracking Operations

Source: https://github.com/midday-ai/midday/blob/main/packages/cli/README.md

Control time tracking, including starting, stopping, and checking timer status. Manage projects for time entries.

```bash
midday tracker start --project proj_abc --description "API development"
```

```bash
midday tracker stop
```

```bash
midday tracker projects create --name "Website Redesign" --rate 150
```

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

### Type Guards and Assertions

Source: https://github.com/midday-ai/midday/blob/main/packages/inbox/README.md

Provides examples of using type guards (`isInboxAuthError`, `isInboxSyncError`) and assertions (`assertInboxAuthError`, `assertInboxSyncError`) for robust error handling.

```APIDOC
### Type Guards and Assertions

- **`isInboxAuthError(error)`**: Type guard that returns `true` if the provided error is an `InboxAuthError`.
- **`isInboxSyncError(error)`**: Type guard that returns `true` if the provided error is an `InboxSyncError`.
- **`assertInboxAuthError(error)`**: Assertion function that narrows the type to `InboxAuthError`. Throws an error if the provided error is not an `InboxAuthError`.
- **`assertInboxSyncError(error)`**: Assertion function that narrows the type to `InboxSyncError`. Throws an error if the provided error is not an `InboxSyncError`.

### Usage Example

```typescript
import {
  isInboxAuthError,
  isInboxSyncError,
  assertInboxAuthError,
} from "@midday/inbox/errors";

// Using type guards
if (isInboxAuthError(error)) {
  // TypeScript knows 'error' is InboxAuthError here
  console.log(error.requiresReauth);
}

// Using assertions
try {
  assertInboxAuthError(error);
  // TypeScript knows 'error' is InboxAuthError here
  console.log(error.provider);
} catch (e) {
  // Handle cases where it's not an InboxAuthError
  if (isInboxSyncError(e)) {
    console.log(e.code);
  }
}
```
```

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

### Environment Variables for Accounting Provider Configuration

Source: https://github.com/midday-ai/midday/blob/main/packages/accounting/README.md

Lists the necessary environment variables for configuring connections to Xero, QuickBooks, and Fortnox, including client IDs, secrets, and redirect URLs. Also includes a variable for OAuth state encryption.

```bash
# Xero
XERO_CLIENT_ID=your_client_id
XERO_CLIENT_SECRET=your_client_secret
XERO_OAUTH_REDIRECT_URL=https://api.midday.ai/v1/apps/xero/oauth-callback

# QuickBooks
QUICKBOOKS_CLIENT_ID=your_client_id
QUICKBOOKS_CLIENT_SECRET=your_client_secret
QUICKBOOKS_OAUTH_REDIRECT_URL=https://api.midday.ai/v1/apps/quickbooks/oauth-callback

# Fortnox
FORTNOX_CLIENT_ID=your_client_id
FORTNOX_CLIENT_SECRET=your_client_secret
FORTNOX_OAUTH_REDIRECT_URL=https://api.midday.ai/v1/apps/fortnox/oauth-callback

# OAuth state encryption
ACCOUNTING_OAUTH_SECRET=32_byte_encryption_key
```

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

### Database Query: Get Transactions for Sync

Source: https://github.com/midday-ai/midday/blob/main/packages/accounting/README.md

Retrieves transactions from the database that are pending synchronization with an accounting provider. Requires database connection and sync parameters.

```typescript
// Get transactions for export
getTransactionsForAccountingSync(db, {
  teamId: string,
  provider: ProviderType,
  transactionIds: string[],  // Required for manual export
  limit?: number,
}): Promise<TransactionForSync[]>
```

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

### Revoke Token Request Example

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/oauth-api-endpoints.mdx

Use this cURL command to revoke an access or refresh token. Provide the token, client ID, and client secret.

```bash
curl -X POST https://api.midday.ai/v1/oauth/revoke \
  -H "Content-Type: application/json" \
  -d '{
    "token": "mid_at_xxxxxxxxxxxxx",
    "client_id": "mid_client_abc123",
    "client_secret": "mid_secret_xyz789"
  }'
```

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

### Create Time Entry

Source: https://context7.com/midday-ai/midday/llms.txt

Create a new time entry with project ID, date, start/stop times, duration, and description. Requires `Authorization` and `Content-Type` headers.

```bash
# Create a time entry
curl -X POST https://api.midday.ai/tracker-entries \
  -H "Authorization: Bearer mid_live_abc123" \
  -H "Content-Type: application/json" \
  -d '{
    "projectId": "proj_uuid",
    "date": "2024-06-20",
    "start": "2024-06-20T09:00:00Z",
    "stop": "2024-06-20T12:00:00Z",
    "duration": 10800,
    "description": "API integration work"
  }'
```

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

### Create and send an invoice

Source: https://context7.com/midday-ai/midday/llms.txt

Create a new invoice and immediately send it to the customer via email. Specify line items, dates, and optional VAT/discount.

```bash
curl -X POST https://api.midday.ai/invoices \
  -H "Authorization: Bearer mid_live_abc123" \
  -H "Content-Type: application/json" \
  -d '{
    "customerId": "cust_uuid",
    "deliveryType": "create_and_send",
    "issueDate": "2024-06-20T00:00:00Z",
    "dueDate": "2024-07-20T00:00:00Z",
    "lineItems": [
      { "name": "Web Development", "quantity": 40, "price": 125.00, "unit": "hours" },
      { "name": "Design Review", "quantity": 5, "price": 100.00, "unit": "hours" }
    ],
    "vat": 250.00,
    "discount": 100.00
  }'
```

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

### Authorization Endpoint Error Response Example

Source: https://github.com/midday-ai/midday/blob/main/apps/website/src/app/docs/content/oauth-api-endpoints.mdx

In case of an error during authorization, the user is redirected to your `redirect_uri` with `error` and `error_description` parameters, along with the original `state`.

```http
https://yourapp.com/callback?error=access_denied&error_description=User%20denied%20access&state=xyz789
```