MCP Server Steel Scraper
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| STEEL_LOCAL | No | Set to 'true' when using a local Steel instance for stateful sessions | |
| STEEL_API_KEY | No | Required for cloud mode stateful sessions | |
| STEEL_API_URL | No | The steel-dev API endpoint (default: http://localhost:3000) | http://localhost:3000 |
| STEEL_RETRIES | No | Number of retry attempts (default: 3) | 3 |
| STEEL_TIMEOUT | No | Request timeout in milliseconds (default: 30000) | 30000 |
| STEEL_BASE_URL | No | Base URL for the Steel Sessions API (default: https://api.steel.dev, or http://localhost:3000 when STEEL_LOCAL=true) | https://api.steel.dev |
| STEEL_IDLE_TIMEOUT_MS | No | Auto-release idle sessions after this many milliseconds (default: 600000) | 600000 |
| STEEL_SESSION_TIMEOUT_MS | No | Session timeout in milliseconds (default: 900000) | 900000 |
| STEEL_GLOBAL_WAIT_SECONDS | No | Optional delay after each stateful action (default: 0) | 0 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| visit_with_browserA | Visit any website using full browser automation (stealth mode, anti-detection). Returns page content in your chosen format: 'html' for raw HTML source, 'markdown' for clean formatted text (recommended for reading), 'readability' for Mozilla Readability format, or 'cleaned_html' for cleaned HTML. Supports screenshot and PDF generation. Automatically handles JavaScript rendering and provides clean output by default. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 1 tool
Only one tool exists, so there is no possibility of confusion with other tools.
With a single tool, naming consistency is inherently perfect as there are no other names to compare against.
The server has only one tool, which is far too few for a robust web scraping service; typically multiple tools would be expected for different formats or actions.
The tool covers visiting a website and returning content in various formats, including screenshots and PDFs, but lacks separate tools for actions like extracting specific data or handling sessions, which might be needed for a full scraping workflow.