### Quick Install and Run

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/OVERVIEW.md

Instructions for quickly installing and running the Deep Research MCP Server using npx. Alternatively, clone the repository, install dependencies, and run the start script.

```bash
npx @pinkpixel/deep-research-mcp
```

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

### Install and Run Deep Research MCP Server

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Provides commands for quick start, global installation, and local development setup of the Deep Research MCP Server. Requires TAVILY_API_KEY.

```bash
TAVILY_API_KEY="tvly-YOUR_KEY" npx @pinkpixel/deep-research-mcp
```

```bash
npm install -g @pinkpixel/deep-research-mcp
TAVILY_API_KEY="tvly-YOUR_KEY" deep-research-mcp
```

```bash
git clone https://github.com/pinkpixel-dev/deep-research-mcp.git
cd deep-research-mcp
npm install
cp .env.example .env          # Add TAVILY_API_KEY to .env
npm run dev                   # nodemon + ts-node
```

```bash
npm run build                 # tsc → dist/
npm start                     # node dist/index.js
```

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

### Install Dependencies for Local Development

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Install project dependencies after cloning the repository for local development.

```bash
npm install
```

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

### Run Server with NPX or Global Install

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Execute the server using npx for direct execution or if the package is installed globally. Ensure environment variables are configured.

```bash
npx @pinkpixel/deep-research-mcp
```

```bash
deep-research-mcp
```

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

### Install and Run (Local)

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/OVERVIEW.md

Instructions for installing and running the Deep Research MCP Server locally after cloning the repository. Requires Node.js >=18.0.0.

```bash
npm install
npm start
```

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

### Build and Run Production Server

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

First, build the TypeScript code for production. Then, start the server using the built JavaScript files.

```bash
npm run build
npm start
```

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

### Install Deep Research MCP Globally

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Optional global installation of the deep-research-mcp package.

```bash
npm install -g @pinkpixel/deep-research-mcp
```

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

### Install Deep Research MCP via Smithery

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Use this command to automatically install the deep-research-mcp for Claude Desktop using Smithery.

```bash
npx -y @smithery/cli install @pinkpixel/deep-research-mcp --client claude
```

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

### File Writing Security Model - Configuration Example

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Configure file writing security by enabling it (`FILE_WRITE_ENABLED=true`), specifying allowed paths (`ALLOWED_WRITE_PATHS`), and setting a line limit (`FILE_WRITE_LINE_LIMIT`).

```bash
# Example: tightly scoped write permissions
FILE_WRITE_ENABLED=true
ALLOWED_WRITE_PATHS=/home/researcher/output,/mnt/nas/reports
FILE_WRITE_LINE_LIMIT=500
```

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

### Run Globally Installed Deep Research MCP

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Command to run the deep-research-mcp after global installation.

```bash
deep-research-mcp
```

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

### MCP Client Configuration for Deep Research Server

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Example JSON configuration for registering the Deep Research MCP Server in an MCP-compatible client, specifying environment variables for server behavior.

```json
{
  "mcpServers": {
    "deep-research": {
      "command": "npx",
      "args": ["-y", "@pinkpixel/deep-research-mcp"],
      "env": {
        "TAVILY_API_KEY": "tvly-YOUR_ACTUAL_API_KEY_HERE",
        "DOCUMENTATION_PROMPT": "Generate a concise technical report with an executive summary, key findings, and references.",
        "SEARCH_TIMEOUT": "120",
        "CRAWL_TIMEOUT": "300",
        "MAX_SEARCH_RESULTS": "10",
        "CRAWL_MAX_DEPTH": "2",
        "CRAWL_LIMIT": "15",
        "FILE_WRITE_ENABLED": "true",
        "ALLOWED_WRITE_PATHS": "/home/user/research,/home/user/documents",
        "FILE_WRITE_LINE_LIMIT": "300"
      }
    }
  }
}
```

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

### Deep Research MCP Server Configuration

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Example JSON configuration for the deep-research-mcp server, including required and optional environment variables.

```json
{
  "mcpServers": {
    "deep-research": {
      "command": "npx",
      "args": [
        "-y",
        "@pinkpixel/deep-research-mcp"
      ],
      "env": {
        "TAVILY_API_KEY": "tvly-YOUR_ACTUAL_API_KEY_HERE", // Required
        "DOCUMENTATION_PROMPT": "Your custom, detailed instructions for the LLM on how to generate markdown documents from the research data...", // Optional - if not provided, the default prompt will be used
        "SEARCH_TIMEOUT": "120", // Optional - timeout in seconds for search requests (default: 60)
        "CRAWL_TIMEOUT": "300", // Optional - timeout in seconds for crawl requests (default: 180)
        "MAX_SEARCH_RESULTS": "10", // Optional - maximum search results to retrieve (default: 7)
        "CRAWL_MAX_DEPTH": "2", // Optional - maximum crawl depth (default: 1)
        "CRAWL_LIMIT": "15", // Optional - maximum URLs to crawl per source (default: 10)
        "FILE_WRITE_ENABLED": "true", // Optional - enable file writing capability (default: false)
        "ALLOWED_WRITE_PATHS": "/home/user/research,/home/user/documents", // Optional - comma-separated allowed directories (default: user home directory)
        "FILE_WRITE_LINE_LIMIT": "300" // Optional - maximum lines per file write operation (default: 200)
      }
    }
  }
}
```

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

### LLM Tool Usage for File Writing

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Example of how an LLM can use the 'write-research-file' tool to save content to a specified file path with a given mode.

```json
{
  "tool": "write-research-file",
  "arguments": {
    "file_path": "/home/user/research/quantum-computing-report.md",
    "content": "# Quantum Computing Research Report\n\n...",
    "mode": "rewrite"
  }
}
```

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

### Enable File Writing and Set Paths

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Enable the secure file writing tool and define allowed write paths and line limits using environment variables.

```env
FILE_WRITE_ENABLED=true
ALLOWED_WRITE_PATHS=/home/user/research,/home/user/documents,/tmp/research
FILE_WRITE_LINE_LIMIT=500
```

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

### Set Custom Documentation Prompt using .env file

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Define a custom prompt for the LLM by setting the DOCUMENTATION_PROMPT variable in a .env file.

```env
DOCUMENTATION_PROMPT="Your custom, detailed instructions for the LLM on how to generate markdown..."
```

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

### Set Custom Prompt and API Key via command line

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Configure both the custom documentation prompt and Tavily API key when executing the npx command.

```bash
DOCUMENTATION_PROMPT="Your custom prompt..." TAVILY_API_KEY="tvly-YOUR_KEY" npx @pinkpixel/deep-research-mcp
```

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

### Configure Performance Settings using .env file

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Adjust search and crawl timeouts, and limits for search results and crawl depth/count by setting environment variables in a .env file.

```env
SEARCH_TIMEOUT=120
CRAWL_TIMEOUT=300
MAX_SEARCH_RESULTS=10
CRAWL_MAX_DEPTH=2
CRAWL_LIMIT=15
```

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

### Set Tavily API Key using .env file

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Create a .env file in the project root and add your Tavily API key.

```env
TAVILY_API_KEY="tvly-YOUR_ACTUAL_API_KEY"
```

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

### Set Output Path and API Key via command line

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Configure the research output path and Tavily API key when running the npx command.

```bash
RESEARCH_OUTPUT_PATH="/path/to/your/research/folder" TAVILY_API_KEY="tvly-YOUR_KEY" npx @pinkpixel/deep-research-mcp
```

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

### Documentation Prompt Precedence - Per-Call Override

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Override the documentation prompt for a specific tool call by including the `documentation_prompt` argument. This takes the highest precedence.

```json
{
  "name": "deep-research-tool",
  "arguments": {
    "query": "...",
    "documentation_prompt": "Create a 5-slide presentation outline with speaker notes."
  }
}
```

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

### Run Development Server with Auto-Reload

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Use this command to run the server in development mode. It utilizes nodemon and ts-node for automatic restarts upon file changes.

```bash
npm run dev
```

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

### Documentation Prompt Precedence - Environment Variable Override

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Override the default documentation prompt for all calls by setting the `DOCUMENTATION_PROMPT` environment variable. This applies to all subsequent calls from the server instance.

```bash
# Override via environment (applies to all calls from this server instance)
export DOCUMENTATION_PROMPT="Generate a concise executive summary followed by detailed technical findings. Use tables where appropriate. Cite all sources inline."
```

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

### Configure Timeouts via command line

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Set search and crawl timeouts directly in the command line when executing the npx command.

```bash
SEARCH_TIMEOUT=120 CRAWL_TIMEOUT=300 TAVILY_API_KEY="tvly-YOUR_KEY" npx @pinkpixel/deep-research-mcp
```

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

### Deep Research MCP Server Environment Variables Reference

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Lists and explains environment variables for configuring the Deep Research MCP Server's behavior, including API keys, search/crawl tuning, and file writing options.

```bash
# Required
TAVILY_API_KEY="tvly-YOUR_KEY"

# Documentation generation
DOCUMENTATION_PROMPT="Your custom LLM instructions for markdown output..."

# Search tuning (defaults shown)
SEARCH_TIMEOUT=60           # seconds before search request times out
MAX_SEARCH_RESULTS=7        # results retrieved (1–20)

# Crawl tuning (defaults shown)
CRAWL_TIMEOUT=180           # seconds before each crawl request times out
CRAWL_MAX_DEPTH=1           # link depth from each root URL (capped at 2)
CRAWL_LIMIT=10              # max URLs processed per root URL (capped at 20)

# File writing (disabled by default)
FILE_WRITE_ENABLED=true
ALLOWED_WRITE_PATHS=/home/user/research,/tmp/research
FILE_WRITE_LINE_LIMIT=300   # max lines per write-research-file call
```

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

### Programmatic Usage of DeepResearchMcpServer (TypeScript/Node.js)

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Instantiate and run the `DeepResearchMcpServer` for embedding in applications. Set required environment variables like API keys and timeouts before instantiation.

```typescript
import { DeepResearchMcpServer } from "@pinkpixel/deep-research-mcp";

// Set required env var before instantiating
process.env.TAVILY_API_KEY = "tvly-YOUR_KEY";
process.env.SEARCH_TIMEOUT = "90";
process.env.CRAWL_TIMEOUT = "240";
process.env.MAX_SEARCH_RESULTS = "5";
process.env.FILE_WRITE_ENABLED = "true";
process.env.ALLOWED_WRITE_PATHS = "/tmp/research";

const server = new DeepResearchMcpServer();

// Connect to stdio transport and start listening for MCP requests
server.run().catch((err) => {
  console.error("Server failed to start:", err);
  process.exit(1);
});
```

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

### General Deep Research Tool Configuration

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

This configuration performs a Tavily search for a specific query and crawls the results. It includes parameters for search depth, topic, result limits, and crawl settings. All parameters except 'query' are optional and can fall back to environment variables or defaults.

```json
{
  "name": "deep-research-tool",
  "arguments": {
    "query": "WebAssembly SIMD performance benchmarks 2024",
    "search_depth": "advanced",
    "topic": "general",
    "max_search_results": 5,
    "chunks_per_source": 3,
    "include_answer": true,
    "include_search_images": false,
    "include_domains_search": ["web.dev", "developer.mozilla.org"],
    "exclude_domains_search": ["pinterest.com"],
    "search_timeout": 90,
    "crawl_max_depth": 1,
    "crawl_max_breadth": 5,
    "crawl_limit": 8,
    "crawl_extract_depth": "advanced",
    "crawl_include_images": true,
    "crawl_select_paths": ["/docs/.*", "/blog/.*"],
    "crawl_exclude_paths": ["/tag/.*", "/author/.*"],
    "crawl_timeout": 240,
    "documentation_prompt": "Write a technical whitepaper. Start with an abstract, then benchmark methodology, results table, analysis, and conclusion."
  }
}
```

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

### Dockerfile for Deep Research MCP Server Runtime Stage

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Specifies the runtime stage of the Dockerfile, copying build artifacts and setting environment variables. API keys and file write paths should be injected at runtime.

```dockerfile
# Runtime stage
FROM node:18-alpine
WORKDIR /app
COPY --from=builder /app/dist ./dist
COPY --from=builder /app/node_modules ./node_modules
COPY package.json ./

ENV NODE_ENV=production
# TAVILY_API_KEY must be injected at runtime (never bake into image)
# FILE_WRITE_ENABLED=true
# ALLOWED_WRITE_PATHS=/data/research

ENTRYPOINT ["node", "dist/index.js"]
```

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

### Running Deep Research MCP Docker Container

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Command to build the Docker image and run the Deep Research MCP container. Environment variables for API key, file write enablement, and allowed paths, along with volume mounts, are configured during runtime.

```bash
docker build -t deep-research-mcp .
docker run --rm -i \
  -e TAVILY_API_KEY="tvly-YOUR_KEY" \
  -e FILE_WRITE_ENABLED="true" \
  -e ALLOWED_WRITE_PATHS="/data/research" \
  -v /host/research:/data/research \
  deep-research-mcp
```

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

### Architecture Diagram

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/OVERVIEW.md

Text-based representation of the Deep Research MCP Server's architecture, illustrating the flow between the LLM/Agent, the server, and external APIs.

```text
[LLM/Agent]
    │
    ▼
[Deep Research MCP Server]
    │   ├─> Tavily Search API
    │   └─> Tavily Crawl API
    ▼
[Aggregated JSON Output + Documentation Instructions]
```

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

### Write Research File - Error Response (Path Not Allowed)

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

An error response for disallowed file paths includes `success: false`, an error message detailing the issue and allowed directories, the attempted file path, and the mode.

```json
{
  "success": false,
  "error": "File writing not allowed to path: /etc/passwd. Must be within one of these directories: /home/user/research",
  "file_path": "/etc/passwd",
  "mode": "rewrite"
}
```

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

### Output Path Resolution - Environment Variable Override

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Set a fixed output directory for all research sessions by exporting the `RESEARCH_OUTPUT_PATH` environment variable. This path will be used unless overridden per-call.

```bash
# Set a fixed output directory for all research sessions
export RESEARCH_OUTPUT_PATH="/home/user/research/current-project"
```

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

### Set Tavily API Key via command line

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Set the TAVILY_API_KEY environment variable directly when running the npx command.

```bash
TAVILY_API_KEY="tvly-YOUR_ACTUAL_API_KEY" npx @pinkpixel/deep-research-mcp
```

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

### Output Path Resolution - Per-Call Override

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Specify an output path for a specific tool call by including the `output_path` argument. This takes precedence over the `RESEARCH_OUTPUT_PATH` environment variable.

```json
{
  "name": "deep-research-tool",
  "arguments": {
    "query": "...",
    "output_path": "/home/user/research/q2-competitive-analysis"
  }
}
```

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

### Write Research File Tool - Rewrite Mode

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Use this tool to save text content to a file. Ensure `FILE_WRITE_ENABLED` is true and the path is allowed. Supports `rewrite` (default) and `append` modes, with a line limit per call.

```json
{
  "name": "write-research-file",
  "arguments": {
    "file_path": "/home/user/research/wasm-simd-report.md",
    "content": "# WebAssembly SIMD Performance Report\n\n## Executive Summary\n\nSIMD instructions in WebAssembly...\n\n## Benchmark Results\n\n| Operation | Scalar (ms) | SIMD (ms) | Speedup |\n|-----------|-------------|-----------|--------|\n| Matrix multiply | 142 | 38 | 3.7× |\n",
    "mode": "rewrite"
  }
}
```

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

### File Structure

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/OVERVIEW.md

The directory structure of the Deep Research MCP Server project, outlining the location of source code, compiled output, assets, and configuration files.

```text
deep-research-mcp/
├── dist/                # Compiled JS output
├── src/
│   └── index.ts         # Main server logic
├── assets/              # Project assets (logo)
├── README.md            # Full documentation
├── OVERVIEW.md          # (You are here!)
├── example_config.json  # Example MCP config
├── package.json         # Project metadata & dependencies
├── tsconfig.json        # TypeScript config
├── CHANGELOG.md         # Version history and changes
```

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

### Set Research Output Path using .env file

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Specify a custom directory for research documents and images by setting RESEARCH_OUTPUT_PATH in a .env file.

```env
RESEARCH_OUTPUT_PATH="/path/to/your/research/folder"
```

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

### Deep Research Tool for Targeted Domain Crawl

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

This configuration restricts crawling to specific documentation subdomains and path patterns, excluding noise from marketing or career pages. It uses `crawl_select_domains` and `crawl_select_paths` for precise targeting.

```json
{
  "name": "deep-research-tool",
  "arguments": {
    "query": "Kubernetes operator pattern best practices",
    "search_depth": "advanced",
    "max_search_results": 6,
    "crawl_max_depth": 2,
    "crawl_max_breadth": 4,
    "crawl_limit": 12,
    "crawl_select_domains": ["^kubernetes\.io$", "^docs\.openshift\.com$"],
    "crawl_select_paths": ["/docs/.*", "/blog/.*"],
    "crawl_exclude_paths": ["/community/.*", "/case-studies/.*"],
    "crawl_allow_external": false,
    "crawl_categories": ["Documentation"],
    "crawl_extract_depth": "advanced",
    "crawl_include_images": false
  }
}
```

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

### Write Research File Tool - Append Mode

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Use this tool to append content to an existing file, particularly useful for reports exceeding the `FILE_WRITE_LINE_LIMIT`. Ensure `FILE_WRITE_ENABLED` is true and the path is allowed.

```json
{
  "name": "write-research-file",
  "arguments": {
    "file_path": "/home/user/research/large-report.md",
    "content": "\n## Section 3: Future Directions\n\nEmerging research indicates...\n",
    "mode": "append"
  }
}
```

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

### Dockerfile for Deep Research MCP Server Build Stage

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Defines the build stage of the Dockerfile for the Deep Research MCP Server, using Node.js 18 Alpine, copying source code, and running npm build.

```dockerfile
# Build stage
FROM node:18-alpine AS builder
WORKDIR /app
COPY package*.json ./ 
RUN npm ci
COPY tsconfig.json ./
COPY src ./src
RUN npm run build
```

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

### Clone Deep Research MCP Repository

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

Clone the repository if you intend to modify or contribute to the deep-research-mcp project.

```bash
git clone [https://github.com/your-username/deep-research-mcp.git](https://github.com/your-username/deep-research-mcp.git)
cd deep-research-mcp
```

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

### Deep Research Tool for News Topics with Date Filter

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

This configuration uses `topic: "news"` with the `days` parameter to limit search results to recent publications. It's useful for competitive intelligence, market monitoring, or tracking fast-moving topics.

```json
{
  "name": "deep-research-tool",
  "arguments": {
    "query": "OpenAI GPT-5 release announcement",
    "topic": "news",
    "days": 7,
    "max_search_results": 8,
    "include_answer": "advanced",
    "search_depth": "advanced",
    "crawl_max_depth": 1,
    "crawl_limit": 5,
    "crawl_extract_depth": "basic",
    "crawl_timeout": 120,
    "documentation_prompt": "Summarize news coverage in bullet points grouped by publication date. Cite each source."
  }
}
```

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

### Deep Research MCP Tool Output Structure

Source: https://github.com/pinkpixel-dev/deep-research-mcp/blob/main/README.md

The deep_research_tool returns a JSON string detailing research findings. This structure includes documentation instructions, original query, search summary, detailed research data from crawled sources, and the output path.

```json
{
  "documentation_instructions": "string", // The detailed prompt for the LLM to generate the markdown.
  "original_query": "string",         // The initial query provided to the tool.
  "search_summary": "string | null",  // An LLM-generated answer/summary from Tavily's search phase (if include_answer was true).
  "research_data": [
    {
      "search_rank": "number",
      "original_url": "string",           // URL of the source found by search.
      "title": "string",                  // Title of the web page.
      "initial_content_snippet": "string",// Content snippet from the initial search result.
      "search_score": "number | undefined",// Relevance score from Tavily search.
      "published_date": "string | undefined",// Publication date (if 'news' topic and available).
      "crawled_data": [
        {
          "url": "string",                // URL of the specific page crawled.
          "raw_content": "string | null", // Rich, extracted content from this page.
          "images": ["string", "..."]     // Array of image URLs found on this page.
        }
      ],
      "crawl_errors": ["string", "..."]   // Array of error messages if crawling this source failed or had issues.
    }
    // ... more sources
  ],
  "output_path": "string"             // Path where research documents and images should be saved.
}
```

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

### Write Research File - Success Response

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

A successful file write operation returns a JSON object indicating success, the file path, mode, content length, and line count.

```json
{
  "success": true,
  "message": "Successfully wrote file: /home/user/research/wasm-simd-report.md",
  "file_path": "/home/user/research/wasm-simd-report.md",
  "mode": "rewrite",
  "content_length": 312,
  "line_count": 8
}
```

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

### Write-Research-File Error Response Structure

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Shows the JSON format for errors returned by the write-research-file tool, including details about the error, file path, and mode.

```json
{
  "success": false,
  "error": "Content exceeds line limit: 450 lines (maximum: 200). Please split content into smaller chunks.",
  "file_path": "/home/user/research/report.md",
  "mode": "rewrite"
}
```

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

### Internal Timeout Pattern for Crawl Operations

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

This TypeScript code demonstrates the internal timeout mechanism using Promise.race for crawl operations. Timeouts can be configured via environment variables or tool arguments.

```typescript
// Internal timeout pattern (from src/index.ts)
const crawlResponse = await Promise.race([
  tavilyClient.crawl(url, crawlParams),
  new Promise((_, reject) =>
    setTimeout(
      () => reject(new Error(`Crawl timeout after ${crawlTimeout}s`)),
      crawlTimeout * 1000
    )
  )
]);

// Tune via environment or per-call argument:
// SEARCH_TIMEOUT=120  (env)  or  search_timeout: 120  (tool arg)
// CRAWL_TIMEOUT=300   (env)  or  crawl_timeout: 300   (tool arg)
```

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

### Override Auto-Restriction with Explicit Domains

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Explicitly define domains to be crawled by providing a JSON configuration. This overrides the default auto-restriction behavior.

```json
{
  "crawl_select_domains": ["^docs\.example\.com$", "^api\.example\.com$"],
  "crawl_allow_external": false
}
```

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

### Deep Research Tool with Time Range Filter

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

This configuration uses the `time_range` parameter to scope search results to a specific recency window without needing to set the topic to 'news'. Accepted values are 'd' (day), 'w' (week), 'm' (month), 'y' (year).

```json
{
  "name": "deep-research-tool",
  "arguments": {
    "query": "Rust async runtime performance comparison",
    "search_depth": "advanced",
    "time_range": "m",
    "max_search_results": 5,
    "include_answer": false,
    "crawl_max_depth": 1,
    "crawl_limit": 8,
    "crawl_extract_depth": "advanced"
  }
}
```

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

### Deep-Research-Tool Error Response Structure

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

Illustrates the JSON structure for error responses from the deep-research-tool. Errors are indicated by `isError: true` in the MCP response.

```json
{
  "documentation_instructions": "<prompt>",
  "error": "Search timeout after 60s",
  "original_query": "original query string"
}
```

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

### Auto-Restriction of Crawl Domains

Source: https://context7.com/pinkpixel-dev/deep-research-mcp/llms.txt

When no specific domains are provided and external crawling is disabled, the server automatically restricts crawls to the hostname of the initial URL. This prevents unintended cross-domain crawling.

```text
URL: https://docs.example.com/api/v2
Auto-generated selectDomains: ["^docs\.example\.com$"]
Effect: Only pages on docs.example.com are followed; links to
        github.com, stackoverflow.com, etc. are ignored.
```

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