Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@nascoder-terminal-browser-mcp Browse https://docs.python.org and show a summary"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
🚀 Ultimate Standalone Terminal Browser & Web Scraper - Browse any website, extract content & links directly in your terminal with zero file pollution. 100% standalone with smart fallbacks, enhanced by optional terminal browsers. Perfect for documentation reading, API exploration & web scraping.
# Install
npm install -g nascoder-terminal-browser-mcp
# Use immediately
browse https://example.com
browse https://docs.python.org --format summary
browse https://news.ycombinator.com --format links# Install
npm install -g nascoder-terminal-browser-mcpEdit ~/.config/amazonq/mcp.json:
{
"mcpServers": {
"nascoder-terminal-browser": {
"command": "npx",
"args": ["nascoder-terminal-browser-mcp"],
"timeout": 30000,
"disabled": false
}
}
}# Exit Q CLI
/quit
# Start again
q chatBrowse https://example.com and show me the contentRelated MCP server: kage-mcp
Feature | Standard Tools | NasCoder Terminal Browser |
File Downloads | ❌ Creates files | ✅ No files - terminal only |
Dependencies | ❌ Requires external tools | ✅ 100% standalone |
Browser Support | Limited | ✅ Multiple engines + fallback |
Fallback Method | None | ✅ Built-in fetch+html-to-text |
Link Extraction | Manual | ✅ Automatic link parsing |
Content Formatting | Raw HTML | ✅ Clean terminal formatting |
Error Handling | Basic | ✅ Advanced retry & fallback |
Output Control | Fixed | ✅ Multiple format options |
Zero external dependencies - Works on any system with Node.js
Built-in fallback - Uses fetch+html-to-text when no terminal browsers available
Smart enhancement - Automatically uses lynx/w3m/links if installed for better formatting
Always functional - Never fails due to missing system tools
No file pollution - Everything displayed directly in terminal
Multiple browser engines - lynx, w3m, links, elinks with auto-selection
Smart fallback - Uses fetch+html-to-text if no terminal browsers available
Clean formatting - Optimized for terminal reading
Link extraction - Automatically find and list all page links
Content truncation - Prevent overwhelming output with length limits
Multiple formats - Choose between full, content-only, links-only, or summary
Error resilience - Multiple fallback methods ensure success
Zero configuration - Works out of the box
Comprehensive logging - Debug issues easily
Flexible options - Customize behavior per request
MCP standard - Integrates with any MCP-compatible system
terminal_browseBrowse websites and display content directly in terminal.
Parameters:
url (required) - Website URL to browse
browser - Terminal browser to use (auto, lynx, w3m, links, elinks)
format - Output format (full, content-only, links-only, summary)
extractLinks - Extract page links (true/false)
maxLength - Maximum content length to prevent overwhelming output
Example:
Use terminal_browse to visit https://docs.github.com with format=summarycheck_browsersCheck which terminal browsers are available on your system.
Example:
Check what terminal browsers are availableextract_linksExtract all links from a webpage without showing full content.
Parameters:
url (required) - Website URL to extract links from
maxLinks - Maximum number of links to return (default: 50)
Example:
Extract all links from https://news.ycombinator.com# Browse any website
browse https://example.com
# Get page summary with stats
browse https://docs.python.org --format summary
# Extract all links
browse https://news.ycombinator.com --format links
# Full content with metadata
browse https://github.com/trending --format full
# Limit content length
browse https://very-long-page.com --max-length 1000
# Use specific browser
browse https://example.com --browser lynxcontent - Clean page text (default)
summary - Brief overview with stats
links - All extracted links
full - Complete content with links
browse <url> [options]
Options:
--format, -f Output format (content, summary, links, full)
--max-length, -l Maximum content length [default: 2000]
--browser, -b Browser to use (auto, lynx, w3m, links)
--help, -h Show helpBrowse https://example.comBrowse https://docs.python.org/3/ with format=content-onlyExtract links from https://github.com/trendingBrowse https://news.ycombinator.com with format=summaryBrowse https://example.com using lynx browserComplete page content
All extracted links
Metadata and statistics
Method used for browsing
Just the page text content
No links or metadata
Clean reading experience
Only the extracted links
Perfect for navigation
Numbered list format
Brief content preview
Key statistics
Quick overview
The package includes everything needed to work:
@modelcontextprotocol/sdk - MCP protocol support
node-fetch - HTTP requests
cheerio - HTML parsing
html-to-text - HTML to text conversion
winston - Logging
For even better text formatting, install terminal browsers:
# macOS (Homebrew)
brew install lynx w3m links
# Ubuntu/Debian
sudo apt install lynx w3m links elinks
# CentOS/RHEL
sudo yum install lynx w3m links elinksFirst Choice: Uses terminal browsers (lynx, w3m, links) if available
Automatic Fallback: Uses built-in fetch+html-to-text if no browsers found
Always Works: Never fails due to missing dependencies
lynx - Best text formatting, recommended
w3m - Good table support, images in some terminals
links - Interactive features, mouse support
elinks - Enhanced links with more features
# macOS (Homebrew)
brew install lynx w3m links
# Ubuntu/Debian
sudo apt install lynx w3m links elinks
# CentOS/RHEL
sudo yum install lynx w3m links elinkslynx (best formatting)
w3m (good compatibility)
links (interactive features)
elinks (enhanced features)
fetch+html-to-text (always available fallback)
{
"url": "https://very-long-page.com",
"maxLength": 5000
}{
"url": "https://documentation-site.com",
"format": "links-only",
"maxLinks": 100
}{
"url": "https://example.com",
"browser": "w3m",
"extractLinks": false
}# Install at least one terminal browser
brew install lynx # macOS
sudo apt install lynx # UbuntuThe tool automatically falls back to fetch+html-to-text
Check internet connectivity
Some sites may block terminal browsers
Use maxLength parameter to limit output:
Browse https://long-page.com with maxLength=2000Check ~/.config/amazonq/mcp.json syntax
Restart Q CLI (/quit then q chat)
Verify package installation: npm list -g nascoder-terminal-browser-mcp
No file system caching (by design)
Memory-efficient processing
Fast response times
Multiple fallback methods
Graceful degradation
Comprehensive error messages
30-second timeout protection
Memory-conscious content truncation
Efficient link extraction
"Finally, a way to browse documentation without cluttering my filesystem with temp files!" - Developer
"The automatic fallback from lynx to fetch+html-to-text saved my workflow when lynx wasn't available." - DevOps Engineer
"Perfect for scraping API docs directly in my terminal. The link extraction is incredibly useful." - API Developer
Tool | Files Created | Browser Support | Link Extraction | Fallback Method |
NasCoder Terminal Browser | ✅ None | ✅ 4 browsers | ✅ Automatic | ✅ fetch+html-to-text |
curl + html2text | ❌ Temp files | ❌ None | ❌ Manual | ❌ None |
wget + pandoc | ❌ Downloads | ❌ None | ❌ Manual | ❌ None |
lynx alone | ❌ Can save files | ✅ lynx only | ❌ Manual | ❌ None |
NPM Package: https://www.npmjs.com/package/nascoder-terminal-browser-mcp
GitHub: https://github.com/freelancernasimofficial/nascoder-terminal-browser-mcp
Issues: https://github.com/freelancernasimofficial/nascoder-terminal-browser-mcp/issues
Author: @freelancernasimofficial
MIT - Feel free to use, modify, and distribute
🚀 Ready to browse the web in your terminal without file clutter?
Install now and experience the difference!
npm install -g nascoder-terminal-browser-mcpBuilt with ❤️ by NasCoder (@freelancernasimofficial)
Check which terminal browsers are available on the system
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must carry the burden of transparency. It clearly indicates a read-only check but does not explain how availability is determined, what browsers are checked, or the return format. There is no contradiction with annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, front-loaded sentence with no wasted words. It conveys the essential purpose efficiently.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the absence of parameters and output schema, the description adequately explains the tool's purpose. However, adding a note about what it returns (e.g., a list of available browsers) would improve completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, which earns a baseline of 4. The description appropriately avoids inventing parameter details since none exist.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('Check') and resource ('terminal browsers available on the system'), clearly distinguishing it from siblings like terminal_browse and extract_links. It unambiguously states what the tool does.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for verifying browser availability before invoking terminal_browse, but it does not explicitly state when to use this tool versus alternatives. It offers no exclusions or prerequisites.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Extract all links from a webpage without displaying the full content
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | URL to extract links from | |
| maxLinks | No | Maximum number of links to return |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden, but it only mentions 'without displaying the full content'. Notably, it says 'all links' while the schema includes a maxLinks parameter defaulting to 50, creating a misleading implication. No details are given about error handling, return format, or any other behavioral traits.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that is front-loaded with the primary action and includes a clarifying clause. Every word earns its place, with no redundant information, making it highly concise and well-structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple two-parameter tool with a complete schema, the description covers the core function but leaves gaps. The 'all links' phrasing conflicts with maxLinks, and there is no mention of return format or behaviors like redirect handling. It is minimally sufficient but not thorough, especially without annotations or an output schema.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3. The description itself does not add meaning beyond the schema; in fact, it conflicts slightly with maxLinks by saying 'all links'. Since the schema already documents both url and maxLinks, the description adds no extra parameter value, so a 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb 'extract' and clearly targets the resource 'all links from a webpage'. It also adds 'without displaying the full content', which differentiates it from sibling tools like terminal_browse. This makes the tool's purpose unambiguous and distinct.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context by noting 'without displaying the full content', suggesting this tool is for when only links are needed, not the full page. However, it does not explicitly name alternatives or state when not to use it, so guidance is only implied rather than explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Browse websites and display content directly in terminal without saving files. Supports multiple terminal browsers (lynx, w3m, links) with fallback to fetch+html-to-text.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | URL to browse (must include http:// or https://) | |
| format | No | Output format | full |
| browser | No | Terminal browser to use (auto selects best available) | auto |
| maxLength | No | Maximum content length to display (prevents overwhelming output) | |
| extractLinks | No | Extract and display links from the page |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the burden of behavioral disclosure. It usefully notes 'without saving files' and the fallback to fetch+html-to-text, which explains error handling. However, it does not disclose return format, error behavior, or side effects beyond these points.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences: the first states the core purpose and key constraint, the second adds browser support/fallback. No wasted words; it's appropriately front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the moderately complex schema (5 params, enums, defaults) and no output schema, the description provides a solid high-level overview. It could mention output formats or the extractLinks feature, but the schema covers those, so the description is adequate for informing tool selection.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema provides 100% coverage of all five parameters with clear descriptions, so the description adds no additional parameter-level detail. Baseline of 3 is appropriate because the schema handles the semantics effectively.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's function: browsing websites and displaying content directly in terminal, with a specific constraint (no file saving) and mention of multiple browser support. It distinguishes itself from sibling tools (check_browsers, extract_links) by focusing on actual browsing/display.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the primary use case (browsing a URL) but does not explicitly contrast with sibling tools or state when not to use it. It offers no guidance on selecting between this and check_browsers/extract_links, though the browser support detail hints at a selection mechanism.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.5check_browsersextract_linksterminal_browseScored across 3 tools
Each tool has a clear, distinct purpose: browsing, checking available browsers, and extracting links. No overlap or ambiguity between them.
All tool names follow a consistent verb_noun pattern (terminal_browse, check_browsers, extract_links), making the set predictable and easy to navigate.
With only 3 tools, the server is tightly scoped to terminal browsing. Each tool is essential and there is no bloat or redundancy.
The tool surface fully covers the stated domain: browsing content, checking browser availability, and extracting links. There are no obvious gaps for typical terminal browsing workflows.
Web scraping for AI agents: scrape, search, crawl, map any website to markdown + JSON. No browser.
Programmatic headless browser sessions: create, interact, browse, and extract - rendered in a real
Extract data from any website with this web scraper tool.
Turn any URL into clean Markdown and structured data. Scrape, crawl, search and extract.