# Mingye Lu Acrawl MCP server

Source: https://glimind.com/server/mcp-registry%2Fio.github.mingye-lu%2Facrawl
Registry id: `mcp-registry/io.github.mingye-lu/acrawl`
Status: unknown

## Tools (44)

- `audit_accessibility`: Run axe-core WCAG accessibility audit
- `cancel_script`: Abort a running script immediately
- `cancel_subagent`: Cancel a running sub-agent by ID
- `click`: Click element by CSS selector, @eN ref, or visible label text
- `click_at`: Click at specific viewport coordinates
- `execute_js`: Run arbitrary JavaScript in page context and return result
- `fill_form`: Fill form fields by selector, name, @eN ref, or visible label text
- `fork`: Spawn a sub-agent on a new browser tab with its own goal and step budget
- `get_page_info`: Read current page URL, title, ready_state; all_tabs lists all tabs
- `get_page_performance`: Get page performance metrics using Navigation and Resource Timing APIs
- `go_back`: Browser back button, returns page_state
- `hover`: Hover over an element to reveal tooltips or menus
- `inspect_cookies`: Inspect cookies with security analysis
- `inspect_log`: Inspect a deduplicated console log group
- `inspect_request`: Inspect a network request by @rN id
- `inspect_storage`: Inspect browser storage (localStorage and sessionStorage)
- `inspect_websocket`: Inspect WebSocket messages for a connection
- `intercept_network`: Manage network interception rules to block or mock requests
- `list_network_activity`: List observed network requests with temporal and URL filtering
- `list_page_logs`: List buffered console logs with level and temporal filtering
- `list_resources`: List all links, images, and forms on current page
- `list_scripts`: List all previously saved scripts with metadata
- `list_websocket_activity`: Overview of WebSocket connections and message counts
- `measure_coverage`: Measure JavaScript and CSS code coverage
- `navigate`: Go to a URL, supports format markdown/text/html/fit_markdown, auto-escalates to browser when JS detected
- `page_map`: Get page's YAML accessibility tree
- `press_key`: Press a keyboard key, optionally targeting an element
- `read_content`: Extract text by heading name or CSS selector with pagination
- `read_script`: Read full JSON definition of a saved script
- `refresh`: Reload current page, returns page_state
- `run_goal`: Execute a high-level crawl goal autonomously with internal LLM loop
- `run_script`: Execute a deterministic multi-step script without per-step LLM calls
- `save_file`: Download a URL to output directory
- `save_script`: Persist a script definition to disk
- `screenshot`: Capture full-page screenshot as base64 PNG
- `script_status`: Check execution status of a running script without blocking
- `scroll`: Scroll up or down by pixel amount
- `select_option`: Select dropdown option by value, label, or index; omit to list options
- `set_device`: Switch browser device emulation (mobile/desktop)
- `subagent_status`: Check status and results of sub-agents without blocking
- `switch_tab`: Switch to a different browser tab by index
- `wait`: Wait for CSS selector state or fixed timeout up to 300s
- `wait_for_scripts`: Block until one or more running scripts complete and return results
- `wait_for_subagents`: Wait for specific or all sub-agents to finish and collect results

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.