Get the relay configuration URL for wet-mcp, opening it in the user's browser when possible. Returns the relay URL, whether the browser launched, and the current status.

Find information across web, academic sources, or library docs. Returns search result listings (titles, URLs, snippets) -- NOT full page content. To read full content from a URL, use the extract tool instead.

Actions:

• search: Web search via SearXNG. Example: search(action="search", query="python async patterns")

• research: Academic/scientific search (Google Scholar, arXiv, PubMed). Example: search(action="research", query="transformer attention mechanism")

• docs: Search library documentation with auto-indexing. Example: search(action="docs", query="how to create routes", library="fastapi")

• docs_resolve: Free-form library name to ranked library_id list. Example: search(action="docs_resolve", query="react")

• docs_query: Version-aware library docs query honoring project lock + token cap. Example: search(action="docs_query", library="react", version="latest", topic="useState", query="how to set initial state")

• docs_lock_project: Detect project manifests (pyproject/package.json/go.mod/Cargo.toml) and lock the library set for Cabinets isolation. Example: search(action="docs_lock_project", project_path="/repo/my-app")

• similar: Find pages similar to a URL (pass URL as query). Example: search(action="similar", query="https://example.com/article")

Key parameters:

• query (required for all actions): Search terms or URL (for similar)

• library (required for docs): Library name, e.g. "react", "fastapi"

• language: Programming language for disambiguation in docs, e.g. "python", "java"

• expand: Enable LLM query expansion for broader coverage (default: false)

• enrich: Fetch actual page content for richer snippets (default: false, adds latency)

• max_results: Number of results (default: 10)

• time_range: Recency filter -- day, week, month, year

• include_domains / exclude_domains: Domain filters

• region (search only): 2-letter ISO 3166-1 country code (e.g. "US", "vn") geo filter. Backends without region support are skipped (named in a warning); a chain with no region-capable backend returns a structured error naming them.

• refine (search only): Opt-in iterative refinement. When results are empty or low-relevance, re-queries up to 2 times with LLM-rewritten terms and returns the best round (adds LLM latency/cost per round).

Use help tool with tool_name="search" for full parameter documentation.

Read and return full page content from URLs or local files. Use this when you have a specific URL and need its content. For finding URLs first, use the search tool instead.

Actions:

• extract: Get clean content from URLs. Example: extract(action="extract", urls=["https://example.com/article"])

• batch: Batch extract with per-domain rate limiting (max 50 URLs). Example: extract(action="batch", urls=["https://a.com/1", "https://b.com/2"])

• crawl: Deep crawl following links from root URLs. Example: extract(action="crawl", urls=["https://docs.example.com"], depth=2)

• map: Discover site URL structure without extracting content. Example: extract(action="map", urls=["https://example.com"])

• convert: Convert local files (PDF, DOCX, PPTX, XLSX) to Markdown. Example: extract(action="convert", paths=["/home/user/report.pdf"])

• extract_structured: Extract structured data using JSON Schema + LLM. Example: extract(action="extract_structured", urls=["https://example.com/pricing"], schema={"type": "object", "properties": {"price": {"type": "string"}}})

• agent: Multi-step research orchestration -- search the web, extract top results, synthesize a cited Markdown answer. Example: extract(action="agent", query="latest pydantic 2 changes", max_urls=5)

• interact: Drive a page with click/fill/submit via patchright. Example: extract(action="interact", url="https://example.com/login", actions=[{"type": "fill", "selector": "#email", "value": "x@y.com"}, {"type": "submit", "selector": "form"}])

• diff: Track content changes across fetches of the same URL(s). Example: extract(action="diff", urls=["https://example.com/pricing"])

Key parameters:

• urls (required for extract/batch/crawl/map/extract_structured/diff): List of URLs

• paths (required for convert): List of local file paths

• query (required for agent): Research question to answer

• url (required for interact): Page URL to drive

• actions (required for interact): List of {type, selector?, description?, value?} ops

• max_urls (agent): Default 5, hard cap 20

• synthesis_model (agent): Override LLM model for the synthesis step

• token_budget (agent): Max prompt tokens (default 10000)

• session (interact): Persistent session id; reuses browser across calls

• screenshot (interact): Capture post-interaction screenshot

• format: Output format -- "markdown" (default), "text", "html"

• depth: Crawl depth (default: 2, max: 5)

• max_pages: Max pages for crawl/map (default: 20, max: 100)

• stealth: Enable anti-bot bypass for protected sites (default: false)

• schema: JSON Schema dict for extract_structured

• refetch (diff): Fetch a fresh copy before comparing (default: true). Set false to compare already-recorded snapshots without a new network fetch.

Use help tool with tool_name="extract" for full parameter documentation.

Discover and download media files (images, videos, audio) from web pages.

Actions:

• list: Scan a page and return media URLs with metadata. Example: media(action="list", url="https://example.com/gallery", media_type="images")

• download: Download media files to local storage. Example: media(action="download", media_urls=["https://example.com/photo.jpg"])

Key parameters:

• url (required for list): Page URL to scan

• media_urls (required for download): List of media URLs to download

• media_type: Filter for list -- "images", "videos", "audio", "files", "all" (default: "all")

• output_dir: Download directory (default: ~/.wet-mcp/downloads)

• prompt: Reserved -- accepted for backward compatibility, ignored

Typical workflow: list (discover) -> download (save locally). For LLM analysis (vision/audio/video), hand the downloaded path to imagine-mcp's understand action. The legacy media(action="analyze") was REMOVED in wet v2.0.0 (deprecated since v1.x.y); calling it now returns the standard unknown-action error.

Use help tool with tool_name="media" for full documentation.

Get detailed documentation for any tool. Call this when you need full parameter reference or usage examples.

Valid tool_name values: search, extract, media, config.

Quick guide -- which tool to use:

• Need to FIND information? Use search (returns result listings with URLs)

• Need to READ a page? Use extract (returns full page content from a URL)

• Need media files? Use media (discover, download images/videos/audio)

• Need server settings? Use config (status, cache, settings, warmup, sync setup)

Server config and management. Actions: status|set|cache_clear|docs_reindex|warmup|setup_sync|setup_status|setup_start|setup_skip|setup_reset|setup_complete. Use help tool with tool_name='config' for full docs.