### Clone and Install Open Accountant CLI

Source: https://github.com/openaccountant/wilson/blob/main/CONTRIBUTING.md

Clone the repository, navigate to the directory, and install dependencies using Bun. This is the initial setup for development.

```bash
git clone https://github.com/open-accountant/cli.git
cd cli
bun install
```

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

### Install and Run Open Accountant CLI

Source: https://github.com/openaccountant/wilson/blob/main/README.md

Follow these steps to clone the repository, install dependencies, configure environment variables, and start the application.

```bash
git clone https://github.com/openaccountant/wilson.git
cd open-accountant
bun install
cp env.example .env   # edit with your API key(s)
bun start
```

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

### Skills System Example

Source: https://context7.com/openaccountant/wilson/llms.txt

Illustrates the structure of a SKILL.md file for defining multi-step workflows. Skills can be invoked via CLI or programmatically.

```typescript
// Skills are invoked via CLI: /skill subscription-audit
// Or programmatically through the skill tool

// Example SKILL.md structure (src/skills/subscription-audit/SKILL.md):
// ---
// name: subscription-audit
// description: Analyzes recurring charges to find subscriptions...
// ---
//
// # Subscription Audit
// ## Workflow
// 1. Use anomaly_detect with types ["unused_subscriptions"]
// 2. Use transaction_search with query "all recurring transactions"
// 3. Calculate monthly/annual costs
// 4. Generate recommendations

// Built-in skills include:
// - subscription-audit: Find and analyze recurring charges
// - tax-prep: Prepare tax deduction summary
// - monthly-digest: Generate monthly financial summary
// - cash-flow-forecast: Project future cash flow
// - net-worth: Calculate and track net worth
// - spending-review: Analyze spending patterns
// - quarterly-taxes: Estimate quarterly tax payments
// - debt-payoff: Create debt payoff strategies
// - And 40+ more specialized financial workflows

// Custom skills can be added to:
// - ~/.openaccountant/skills/  (user-wide)
// - .openaccountant/skills/    (project-specific)
```

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

### Initialize and Get Net Worth Summary

Source: https://context7.com/openaccountant/wilson/llms.txt

Initializes the net worth tool and retrieves the current net worth summary. Requires a database connection for initialization.

```typescript
import { netWorthTool, initNetWorthTool } from './src/tools/net-worth/net-worth.js';

initNetWorthTool(db);

// Get current net worth summary
const summary = await netWorthTool.func({
  action: 'summary'
});
```

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

### Get CLI Version

Source: https://github.com/openaccountant/wilson/blob/main/CONTRIBUTING.md

Command to retrieve the current version of the Open Accountant CLI. Useful for bug reporting.

```bash
wilson --version
```

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

### Run Development Commands for Open Accountant CLI

Source: https://github.com/openaccountant/wilson/blob/main/CONTRIBUTING.md

Commands to run tests, perform type checking, and start the development server for the Open Accountant CLI project.

```bash
bun test
bun run typecheck
bun run dev
```

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

### Conventional Commit Message Examples

Source: https://github.com/openaccountant/wilson/blob/main/CONTRIBUTING.md

Examples of commit messages following the conventional commit format, indicating the type of change (feat, fix, docs, refactor, test).

```git
feat: add new bank parser for Wells Fargo
fix: resolve CSV parsing edge case
docs: update installation instructions
refactor: simplify transaction matching
test: add tests for OFX parser
```

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

### Get Full Balance Sheet

Source: https://context7.com/openaccountant/wilson/llms.txt

Generates a full balance sheet report using the net worth tool.

```typescript
// Get full balance sheet
const balanceSheet = await netWorthTool.func({
  action: 'balance_sheet'
});
```

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

### Get Net Worth Trend

Source: https://context7.com/openaccountant/wilson/llms.txt

Retrieves the net worth trend over a specified number of months. This is a Pro feature.

```typescript
// Get net worth trend (Pro feature)
const trend = await netWorthTool.func({
  action: 'trend',
  months: 12
});
```

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

### Get Tax Deduction Summary

Source: https://context7.com/openaccountant/wilson/llms.txt

Retrieves a summary of tax-deductible transactions for a specified tax year, including category totals and a grand total.

```typescript
// Get tax deduction summary
const summary = await taxFlagTool.func({
  action: 'summary',
  taxYear: 2024
});
```

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

### Create and Run Agent

Source: https://context7.com/openaccountant/wilson/llms.txt

Instantiate an Agent with a configured model and run it with a query. The agent processes events like thinking, tool usage, and final results.

```typescript
import { Agent } from './src/agent/agent.js';

// Create agent with configured model
const agent = await Agent.create({
  model: 'gpt-4o',           // or 'claude-3-5-sonnet', 'ollama:llama3.2', etc.
  maxIterations: 10,
  signal: abortController.signal
});

// Run the agent with a query
for await (const event of agent.run('What did I spend on dining last month?')) {
  switch (event.type) {
    case 'thinking':
      console.log('Agent thinking:', event.message);
      break;
    case 'tool_start':
      console.log(`Calling ${event.tool}...`);
      break;
    case 'tool_result':
      console.log(`${event.tool} completed:`, event.result);
      break;
    case 'done':
      console.log('Final answer:', event.answer);
      console.log('Token usage:', event.tokenUsage);
      break;
  }
}
```

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

### Initialize and Sync Plaid Bank Accounts

Source: https://context7.com/openaccountant/wilson/llms.txt

Initializes the Plaid sync tool and synchronizes all linked bank accounts. This process automatically creates accounts and updates balances.

```typescript
import { plaidSyncTool, initPlaidSyncTool } from './src/tools/import/plaid-sync.js';

initPlaidSyncTool(db);

// Sync all linked bank accounts
const result = await plaidSyncTool.func({});
```

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

### Configure MCP Servers

Source: https://github.com/openaccountant/wilson/blob/main/README.md

Add external tool servers by defining their commands and arguments in the MCP configuration file. This allows Open Accountant to integrate with custom tools.

```json
{
  "servers": {
    "my-server": {
      "command": "npx",
      "args": ["-y", "my-mcp-server"],
      "env": {}
    }
  }
}
```

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

### Open Accountant CLI Project Structure

Source: https://github.com/openaccountant/wilson/blob/main/CONTRIBUTING.md

Overview of the directory structure for the Open Accountant CLI project, highlighting key directories like agent, tools, and skills.

```tree
src/
├── agent/         # Agent loop and orchestration
├── tools/         # Tool registry and implementations
│   ├── import/    # Bank parsers (Chase, Amex, etc.)
│   └── ...
├── skills/        # Built-in skill definitions
├── providers.ts   # LLM provider routing
└── index.ts       # CLI entry point
```

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

### Manage Monthly Budgets

Source: https://context7.com/openaccountant/wilson/llms.txt

Set spending limits per category and check current budget status using the budget_set and budget_check tools.

```typescript
import { budgetSetTool, initBudgetSetTool } from './src/tools/budget/budget-set.js';
import { budgetCheckTool, initBudgetCheckTool } from './src/tools/budget/budget-check.js';

initBudgetSetTool(db);
initBudgetCheckTool(db);

// Set monthly budget limits
await budgetSetTool.func({ category: 'Dining', monthlyLimit: 400 });
await budgetSetTool.func({ category: 'Groceries', monthlyLimit: 800 });
await budgetSetTool.func({ category: 'Shopping', monthlyLimit: 300 });

// Check budget status for current month
const result = await budgetCheckTool.func({});

// Result:
// {
//   month: '2024-01',
//   budgets: [
//     { category: 'Dining', budget: 400, actual: 325.50, remaining: 74.50, percentUsed: 81.4, status: 'OK' },
//     { category: 'Groceries', budget: 800, actual: 745.00, remaining: 55.00, percentUsed: 93.1, status: 'OK' },
//     { category: 'Shopping', budget: 300, actual: 412.00, remaining: -112.00, percentUsed: 137.3, status: 'OVER' }
//   ]
// }

// Check specific category for specific month
const diningCheck = await budgetCheckTool.func({
  month: '2024-01',
  category: 'Dining'
});
```

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

### Generate Spending Summaries

Source: https://context7.com/openaccountant/wilson/llms.txt

The spendingSummaryTool generates spending breakdowns by category for a specified period, with an option to compare with the previous period. Initialize the tool with initSpendingSummaryTool(db) before use.

```typescript
import { spendingSummaryTool, initSpendingSummaryTool } from './src/tools/query/spending-summary.js';

initSpendingSummaryTool(db);

// Get monthly spending breakdown with comparison
const result = await spendingSummaryTool.func({
  period: 'month',
  compareWithPrevious: true
});

// Result:
// {
//   period: 'January 2024',
//   dateRange: { start: '2024-01-01', end: '2024-01-31' },
//   totalSpending: -3425.67,
//   transactionCount: 156,
//   categories: [
//     { category: 'Groceries', total: -845.32, count: 12 },
//     { category: 'Dining', total: -523.45, count: 28 },
//     { category: 'Shopping', total: -412.00, count: 15 },
//     ...
//   ],
//   previousPeriod: {
//     label: 'December 2023',
//     totalSpending: -3102.45,
//     categories: [...] 
//   },
//   formatted: 'Spending Summary: January 2024\n\nCategory            Amount       Count  Change\n...'
// }

// Quarterly summary without comparison
const quarterlyResult = await spendingSummaryTool.func({
  period: 'quarter',
  compareWithPrevious: false
});
```

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

### Initialize and Flag Transaction for Tax Deductions

Source: https://context7.com/openaccountant/wilson/llms.txt

Initializes the tax flag tool and flags a specific transaction as tax-deductible, assigning an IRS category and notes. Requires a database connection for initialization.

```typescript
import { taxFlagTool, initTaxFlagTool } from './src/tools/tax/tax-flag.js';

initTaxFlagTool(db);

// Flag a transaction as deductible
await taxFlagTool.func({
  action: 'flag',
  transactionId: 123,
  irsCategory: 'Office expenses',
  taxYear: 2024,
  notes: 'Desk for home office'
});
```

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

### Initialize and Import Monarch Money Transactions

Source: https://context7.com/openaccountant/wilson/llms.txt

Initializes the Monarch Money import tool and imports transactions within a specified date range. Requires environment variables for authentication.

```typescript
import { monarchImportTool, initMonarchTool } from './src/tools/import/monarch.js';

initMonarchTool(db);

// Import transactions from Monarch Money
// Requires MONARCH_TOKEN or MONARCH_EMAIL + MONARCH_PASSWORD env vars
const result = await monarchImportTool.func({
  limit: 500,
  startDate: '2024-01-01',
  endDate: '2024-01-31'
});
```

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

### Call LLM with Tool Definitions

Source: https://context7.com/openaccountant/wilson/llms.txt

Integrate tool calling capabilities by providing tool definitions to the callLlm function. The response will include any tool calls made by the LLM.

```typescript
import { callLlm } from './src/model/llm.js';

// With tool definitions
const toolResult = await callLlm('Search for transactions over $100', {
  model: 'gpt-4o',
  tools: [transactionSearchTool.definition],
  systemPrompt: 'Use the available tools to help the user.'
});
console.log(toolResult.response.toolCalls); // Array of tool calls
```

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

### Import Bank Transactions with csvImportTool

Source: https://context7.com/openaccountant/wilson/llms.txt

Use `csvImportTool` to import transactions from CSV, OFX, or QIF files. Supports auto-detection of bank and deduplication. Can import single files or entire directories.

```typescript
import { csvImportTool, initImportTool } from './src/tools/import/csv-import.js';
import { Database } from 'better-sqlite3';

// Initialize database and tool
const db = new Database('~/.openaccountant/data.db');
initImportTool(db);

// Import a single Chase CSV file
const result = await csvImportTool.func({
  filePath: '~/Downloads/chase-transactions.csv',
  bank: 'auto' // Auto-detects Chase, Amex, BofA, or generic
});

// Result:
// {
//   success: true,
//   transactionsImported: 156,
//   transactionsSkipped: 0,
//   transactionsLinked: 42,
//   formatDetected: 'csv',
//   bankDetected: 'chase',
//   dateRange: { start: '2024-01-01', end: '2024-01-31' },
//   message: 'Imported 156 transactions from chase CSV (2024-01-01 to 2024-01-31). 42 transactions auto-linked to accounts.'
// }

// Import entire directory of bank files
const dirResult = await csvImportTool.func({
  filePath: '~/Downloads/bank-statements/'
});

// Result:
// {
//   success: true,
//   directory: '~/Downloads/bank-statements/',
//   filesFound: 4,
//   filesImported: 3,
//   filesSkipped: 1,
//   totalTransactionsImported: 423,
//   totalTransactionsSkipped: 45,
//   message: 'Directory: ~/Downloads/bank-statements/. Files found: 4, imported: 3, skipped: 1. Transactions imported: 423, skipped: 45.'
// }
```

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

### Skill Marketplace Card Layout

Source: https://github.com/openaccountant/wilson/blob/main/BRAND.md

Visual structure for a purchasable skill card using a slate background and money-green price badge.

```text
┌─────────────────────────────────┐  surface-2
│  Subscription Audit             │  ledger, Space Grotesk 600
│  Find forgotten charges and     │  ink, Inter 400
│  recurring fees you don't use.  │
│                                 │
│  [$15/yr]          [Activate]   │  money green badge + CTA
└─────────────────────────────────┘  carbon border
```

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

### POST /spending-summary

Source: https://context7.com/openaccountant/wilson/llms.txt

Generates spending breakdowns by category for a specified period with optional period-over-period comparison.

```APIDOC
## POST /spending-summary

### Description
Generates spending breakdowns by category for a given period with optional comparison to the previous period.

### Method
POST

### Request Body
- **period** (string) - Required - The time period for the summary (e.g., 'month', 'quarter')
- **compareWithPrevious** (boolean) - Optional - Whether to include comparison data with the previous period

### Response
#### Success Response (200)
- **period** (string) - The label for the period
- **dateRange** (object) - Start and end dates
- **totalSpending** (number) - Total amount spent
- **transactionCount** (number) - Total number of transactions
- **categories** (array) - Breakdown of spending by category
- **previousPeriod** (object) - Optional comparison data
```

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

### Create a Feature Branch

Source: https://github.com/openaccountant/wilson/blob/main/CONTRIBUTING.md

Command to create a new Git branch for developing a feature, following the convention 'feature/your-feature-name'.

```bash
git checkout -b feature/amazing-feature
```

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

### Perform Web Search using Exa

Source: https://context7.com/openaccountant/wilson/llms.txt

Performs a web search using the Exa tool. Requires the EXASEARCH_API_KEY environment variable to be set.

```typescript
import { exaSearch } from './src/tools/search/exa.js';

// Requires EXASEARCH_API_KEY env var
const result = await exaSearch.func({
  query: 'is Costco membership worth it for single person'
});
```

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

### Manage Financial Accounts

Source: https://context7.com/openaccountant/wilson/llms.txt

Track various account types for net worth calculations using the account_manage tool.

```typescript
import { accountManageTool, initAccountManageTool } from './src/tools/net-worth/account-manage.js';

initAccountManageTool(db);

// Add accounts
await accountManageTool.func({
  action: 'add',
  name: 'Chase Checking',
  accountSubtype: 'checking',
  institution: 'Chase',
  accountNumberLast4: '4521',
  currentBalance: 5432.10
});

await accountManageTool.func({
  action: 'add',
  name: 'Fidelity 401k',
  accountSubtype: 'retirement_401k',
  institution: 'Fidelity',
  currentBalance: 125000.00
});

await accountManageTool.func({
  action: 'add',
  name: 'Chase Sapphire',
  accountSubtype: 'credit_card',
  institution: 'Chase',
  accountNumberLast4: '9876',
  currentBalance: 2345.67  // Always positive, stored as liability
});

// List all accounts
const accounts = await accountManageTool.func({
  action: 'list'
});

// Result:
// {
//   count: 3,
//   accounts: [
//     { id: 1, name: 'Chase Checking', type: 'asset', subtype: 'checking', institution: 'Chase', balance: 5432.10 },
//     { id: 2, name: 'Fidelity 401k', type: 'asset', subtype: 'retirement_401k', institution: 'Fidelity', balance: 125000.00 },
//     { id: 3, name: 'Chase Sapphire', type: 'liability', subtype: 'credit_card', institution: 'Chase', balance: 2345.67 }
//   ]
// }
```

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

### Query Transactions with Natural Language

Source: https://context7.com/openaccountant/wilson/llms.txt

Use the transactionSearchTool to query transactions using natural language. It parses queries into filters for dates, categories, amounts, merchants, and recurring status. Ensure initTransactionSearchTool is called with the database instance before use.

```typescript
import { transactionSearchTool, initTransactionSearchTool } from './src/tools/query/transaction-search.js';

initTransactionSearchTool(db);

// Natural language queries
const diningResult = await transactionSearchTool.func({
  query: 'dining in January'
});
// Parses to: { category: 'Dining', dateStart: '2024-01-01', dateEnd: '2024-01-31' }

const expensiveResult = await transactionSearchTool.func({
  query: 'Amazon purchases over $100 last month'
});
// Parses to: { merchant: 'Amazon', maxAmount: -100, dateStart: '...', dateEnd: '...' }

const recurringResult = await transactionSearchTool.func({
  query: 'recurring subscriptions this year'
});
// Parses to: { isRecurring: true, dateStart: '2024-01-01', dateEnd: '2024-12-31' }

// Result structure:
// {
//   query: 'dining in January',
//   filtersApplied: { category: 'Dining', dateStart: '2024-01-01', dateEnd: '2024-01-31' },
//   count: 28,
//   formatted: 'Found 28 transactions:\n\nDate        Amount      Category          Description\n...',
//   transactions: [
//     { id: 1, date: '2024-01-15', description: 'CHIPOTLE', amount: -12.50, category: 'Dining' },
//     ...
//   ]
// }
```

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

### Detect Financial Anomalies

Source: https://context7.com/openaccountant/wilson/llms.txt

Use the anomalyDetectTool to identify financial anomalies such as duplicate charges, spending spikes, and unused subscriptions. Initialize the tool with initAnomalyDetectTool(db) before making calls. Specify anomaly types to detect, or use 'all' for comprehensive detection.

```typescript
import { anomalyDetectTool, initAnomalyDetectTool } from './src/tools/query/anomaly-detect.js';

initAnomalyDetectTool(db);

// Detect all anomaly types
const result = await anomalyDetectTool.func({
  types: ['all']
});

// Result:
// {
//   anomalyCount: 7,
//   duplicates: 2,
//   spikes: 3,
//   unusedSubscriptions: 2,
//   anomalies: [
//     {
//       type: 'duplicate',
//       transactions: [
//         { id: 45, date: '2024-01-15', description: 'SPOTIFY', amount: -9.99 },
//         { id: 46, date: '2024-01-15', description: 'SPOTIFY', amount: -9.99 }
//       ],
//       message: 'Possible duplicate: "SPOTIFY" for $9.99 on 2024-01-15 and 2024-01-15'
//     },
//     {
//       type: 'spike',
//       transaction: { id: 78, date: '2024-01-20', description: 'AMAZON', amount: -456.78 },
//       averageAmount: -45.50,
//       multiplier: 10.0,
//       message: 'Spending spike: "AMAZON" charged $456.78 on 2024-01-20 (10.0x the average of $45.50)'
//     },
//     {
//       type: 'unused_subscription',
//       merchant: 'ADOBE CREATIVE',
//       lastCharge: { id: 92, date: '2024-01-01', amount: -54.99 },
//       daysSinceNonRecurring: 180,
//       message: 'Potentially unused subscription: "ADOBE CREATIVE" ($54.99/charge). No related activity in 180 days.'
//     }
//   ],
//   formatted: 'Anomaly Detection Report\nFound 7 anomalies:\n\nPOTENTIAL DUPLICATES (2)\n...'
// }

// Detect only spending spikes
const spikesOnly = await anomalyDetectTool.func({
  types: ['spikes']
});
```

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

### POST /anomaly-detect

Source: https://context7.com/openaccountant/wilson/llms.txt

Identifies financial anomalies such as duplicate charges, spending spikes, and potentially unused subscriptions.

```APIDOC
## POST /anomaly-detect

### Description
Identifies financial anomalies including duplicate charges, spending spikes, and potentially unused subscriptions.

### Method
POST

### Request Body
- **types** (array) - Required - List of anomaly types to detect (e.g., ['all'], ['spikes'])

### Response
#### Success Response (200)
- **anomalyCount** (number) - Total number of anomalies found
- **duplicates** (number) - Count of duplicate charges
- **spikes** (number) - Count of spending spikes
- **unusedSubscriptions** (number) - Count of potentially unused subscriptions
- **anomalies** (array) - Detailed list of anomaly objects
```

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

### Categorize Transactions with AI using categorizeTool

Source: https://context7.com/openaccountant/wilson/llms.txt

Use `categorizeTool` to classify uncategorized transactions using AI. It assigns transactions to 18 spending categories and provides confidence scores. Can categorize all transactions or a limited set, optionally assigning to a specific entity.

```typescript
import { categorizeTool, initCategorizeTool } from './src/tools/categorize/categorize.js';

initCategorizeTool(db);

// Categorize all uncategorized transactions
const result = await categorizeTool.func({});

// Result:
// {
//   success: true,
//   totalUncategorized: 156,
//   categorized: 156,
//   ruleMatched: 23,           // Pre-categorized by rules engine
//   llmCategorized: 133,       // Categorized by AI
//   categoriesApplied: {
//     Dining: 28,
//     Groceries: 35,
//     Transport: 12,
//     Shopping: 45,
//     Subscriptions: 8,
//     Utilities: 6,
//     Other: 22
//   },
//   needingReview: 15,         // Low confidence (<0.7)
//   message: 'Categorized 156 of 156 transactions (23 by rules, 133 by LLM). 15 need review (confidence < 0.7).'
// }

// Categorize with a limit for specific entity
const limitedResult = await categorizeTool.func({
  limit: 50,
  entityId: 2  // Assign to business entity
});
```

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

### Call LLM with Zod Schema for Structured Output

Source: https://context7.com/openaccountant/wilson/llms.txt

Call the LLM with a Zod schema to ensure structured JSON output. The response will be typed according to the provided schema.

```typescript
import { callLlm } from './src/model/llm.js';
import { z } from 'zod';

// Structured output with Zod schema
const categorySchema = z.object({
  transactions: z.array(z.object({
    id: z.number(),
    category: z.string(),
    confidence: z.number()
  }))
});

const categorization = await callLlm(
  'Categorize these transactions: [...]',
  {
    model: 'claude-3-5-sonnet',
    systemPrompt: 'Respond only with valid JSON.',
    outputSchema: categorySchema
  }
);
console.log(categorization.response.structured); // Typed result
```

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

### Configure Categorization Rules

Source: https://context7.com/openaccountant/wilson/llms.txt

Define glob or regex patterns to auto-categorize transactions. Rules are managed via the rule_manage tool.

```typescript
import { ruleManageTool, initRuleManageTool } from './src/tools/rules/rule-manage.js';

initRuleManageTool(db);

// Add categorization rules (glob patterns by default)
await ruleManageTool.func({
  action: 'add',
  pattern: '*STARBUCKS*',
  category: 'Dining',
  priority: 10
});

await ruleManageTool.func({
  action: 'add',
  pattern: '*NETFLIX*',
  category: 'Subscriptions',
  priority: 10
});

// Add regex-based rule
await ruleManageTool.func({
  action: 'add',
  pattern: '^UBER\\s+(TRIP|EATS)',
  category: 'Transport',
  priority: 5,
  is_regex: true
});

// List all rules
const rules = await ruleManageTool.func({ action: 'list' });

// Result:
// {
//   rules: [
//     { id: 1, pattern: '*STARBUCKS*', category: 'Dining', priority: 10, is_regex: false },
//     { id: 2, pattern: '*NETFLIX*', category: 'Subscriptions', priority: 10, is_regex: false },
//     { id: 3, pattern: '^UBER\\s+(TRIP|EATS)', category: 'Transport', priority: 5, is_regex: true }
//   ],
//   formatted: '#1 [pri:10] "*STARBUCKS*" → Dining\n#2 [pri:10] "*NETFLIX*" → Subscriptions\n#3 [pri:5] "^UBER\\s+(TRIP|EATS)" → Transport (regex)'
// }
```

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

### POST /transaction-search

Source: https://context7.com/openaccountant/wilson/llms.txt

Parses natural language queries into structured filters to search and retrieve financial transactions.

```APIDOC
## POST /transaction-search

### Description
Parses natural language queries into filters for date ranges, categories, amounts, merchants, and recurring status.

### Method
POST

### Request Body
- **query** (string) - Required - The natural language query string (e.g., 'dining in January')

### Response
#### Success Response (200)
- **query** (string) - The original query
- **filtersApplied** (object) - The parsed filter criteria
- **count** (number) - Number of transactions found
- **formatted** (string) - Human-readable summary
- **transactions** (array) - List of matching transaction objects
```

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

### Call LLM for Text Generation

Source: https://context7.com/openaccountant/wilson/llms.txt

Use the callLlm function for simple text generation with a specified model and system prompt. Ensure the model is correctly configured.

```typescript
import { callLlm } from './src/model/llm.js';

// Simple text generation
const result = await callLlm('Summarize my spending in one sentence', {
  model: 'gpt-4o',
  systemPrompt: 'You are a helpful financial assistant.'
});
console.log(result.response.content);
```

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

### Export Transactions to Files

Source: https://context7.com/openaccountant/wilson/llms.txt

Use the export_transactions tool to save filtered transaction data to CSV or XLSX formats. Requires initialization with a database instance before use.

```typescript
import { exportTransactionsTool, initExportTool } from './src/tools/export/export-transactions.js';

initExportTool(db);

// Export all transactions to XLSX
const result = await exportTransactionsTool.func({
  format: 'xlsx',
  filePath: '~/Documents/transactions-2024.xlsx'
});

// Export filtered transactions to CSV
const filteredResult = await exportTransactionsTool.func({
  format: 'csv',
  filePath: '~/Documents/dining-q1.csv',
  dateStart: '2024-01-01',
  dateEnd: '2024-03-31',
  category: 'Dining'
});

// Result:
// {
//   success: true,
//   transactionsExported: 84,
//   format: 'csv',
//   filePath: '/Users/name/Documents/dining-q1.csv',
//   message: 'Exported 84 transactions to /Users/name/Documents/dining-q1.csv (CSV).'
// }
```

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

### List Tax Deductions in a Category

Source: https://context7.com/openaccountant/wilson/llms.txt

Lists all flagged tax deductions within a specific IRS category for a given tax year.

```typescript
// List deductions in a category
const officeDeductions = await taxFlagTool.func({
  action: 'list',
  taxYear: 2024,
  irsCategory: 'Office expenses'
});
```

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

### Transaction Row Layout

Source: https://github.com/openaccountant/wilson/blob/main/BRAND.md

Tabular display format for financial transactions using monospaced fonts and color-coded amounts.

```text
  Mar 01   NETFLIX.COM        Entertainment   -$15.99    ink / mono
  Mar 01   WHOLE FOODS #123   Groceries       -$87.42    ink / mono
  Feb 28   PAYROLL DEPOSIT    Income        +$3,200.00   money green / mono
```

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

### Payment Required State Layout

Source: https://github.com/openaccountant/wilson/blob/main/BRAND.md

Visual structure for a payment-gated skill, featuring a caution-amber header and money-green call-to-action.

```text
┌─────────────────────────────────┐  surface-2
│  ⚠ Payment Required             │  caution amber
│                                 │
│  Tax Prep Skill — $25/year      │  ledger + money green
│  Categorize deductions, flag    │
│  write-offs, export for filing. │  ink
│                                 │
│  [Pay with x402]    [Details]   │  money green CTA
└─────────────────────────────────┘
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.