@webcrawl/mcp 1.3.8 → 1.3.10

package/README.md

@@ -1,8 +1,8 @@
 # @webcrawl/mcp
 
-Model Context Protocol (MCP) server for **[Webcrawl Anything](https://github.com/brainfish-ai/webcrawl-anything)** and **[webcrawl.app](https://webcrawl.app)**.
+Model Context Protocol (MCP) server for **[Webcrawl Anything](https://github.com/brainfish-ai/webcrawl-anything)**.
 
-Gives Claude Code, Cursor, Codex, Windsurf, and Gemini CLI live access to web search, structured verticals (shopping, flights, hotels, real-estate, local business, CVEs), anti-bot page scraping, PDF/Office OCR, and YouTube transcripts.
+Gives Claude Code, Cursor, Codex, Windsurf, and Gemini CLI live access to web search, structured verticals (shopping, flights, hotels, real-estate, local business, CVEs), page reading for JavaScript-heavy sites, and PDF/Office parsing. It runs on your machine with no API key and no account.
 
 ## One-Command Agent Setup
 
@@ -23,14 +23,22 @@ Add to `.cursor/mcp.json`, `~/.claude.json`, or your MCP client config:
       "command": "npx",
       "args": ["-y", "@webcrawl/mcp@latest"],
       "env": {
-        "WEBCRAWL_MCP_SCHEMA_PROFILE": "compact",
-        "WEBCRAWL_API_KEY": "wc_live_..."
+        "WEBCRAWL_MCP_SCHEMA_PROFILE": "compact"
       }
     }
   }
 }
 ```
 
+### Optional: webcrawl.app cloud
+
+Local mode is the default and needs nothing else. To run the same tools on the hosted service, either:
+
+- add `"WEBCRAWL_API_KEY": "wc_live_..."` to `env` above (get a key at [webcrawl.app](https://webcrawl.app)), or
+- connect your client to the hosted endpoint `https://webcrawl.app/mcp` and sign in in the browser. No local install is needed.
+
+`WEBCRAWL_MODE` can be `local`, `cloud` or `auto` (default: local unless a key is set).
+
 ### Compact Schema Profile
 
 Set `WEBCRAWL_MCP_SCHEMA_PROFILE=compact` (default when installed via `npx webcrawl setup`) to minimize tool schema token usage in context-constrained agents while keeping full server-side validation. Set `WEBCRAWL_MCP_SCHEMA_PROFILE=full` for verbose parameter descriptions.
@@ -44,6 +52,7 @@ Set `WEBCRAWL_MCP_SCHEMA_PROFILE=compact` (default when installed via `npx webcr
 - `web_crawl` — Crawl up to 25 pages starting from a URL.
 - `web_extract` — Extract structured JSON from a URL using a prompt or JSON Schema.
 - `web_rerank` — Rerank candidate documents by relevance to a query.
+- `web_monitor` — Watch a page, URL list or site on a schedule with trigger conditions and email/webhook alerts (cloud only).
 - `search_categories` — List all supported search categories, subcategories, and filters.
 
 ## License

package/dist/server.js

@@ -54,7 +54,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             sites: desc(z.array(z.string()).optional(), "Only these domains, e.g. [\"booking.com\"]"),
             include_content: desc(z.boolean().optional(), "Also fetch page markdown for each result (slower, more credits)"),
         },
-        annotations: { readOnlyHint: true, openWorldHint: true },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: true },
     }, ({ query, category, subcategory, limit, country, city, sites, include_content }) => run(async () => {
         const res = await client.search({
             query,
@@ -77,7 +77,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             url: z.string().url(),
             no_cache: desc(z.boolean().optional(), "Skip cached copy"),
         },
-        annotations: { readOnlyHint: true, openWorldHint: true },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: true },
     }, ({ url, no_cache }) => run(async () => {
         const res = await client.scrape({ url, formats: ["markdown"], noCache: no_cache });
         return formatScrape(res) + formatCredits(res);
@@ -94,7 +94,8 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             language: desc(z.string().max(8).optional(), "Spoken language hint, e.g. en"),
             background: desc(z.boolean().optional(), "Run as a background job for long files; returns a run id"),
         },
-        annotations: { readOnlyHint: true, openWorldHint: true },
+        // Not read-only: `background` queues a stateful job. Additive, so not destructive.
+        annotations: { readOnlyHint: false, destructiveHint: false, openWorldHint: true },
     }, ({ url, ocr, pages, language, background }) => run(async () => {
         const res = await client.parse({ url, ocr, pages, language, async: background });
         return formatParse(res) + formatCredits(res);
@@ -103,7 +104,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
         title: "Check a file read",
         description: "Check a background web_parse run and return its Markdown when it has finished.",
         inputSchema: { id: z.string().min(1) },
-        annotations: { readOnlyHint: true, openWorldHint: false },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: false },
     }, ({ id }) => run(async () => {
         const res = await client.run(id);
         const result = res.result && typeof res.result === "object"
@@ -124,7 +125,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             search: desc(z.string().optional(), "Only URLs matching this text"),
             limit: z.number().int().min(1).max(5000).optional(),
         },
-        annotations: { readOnlyHint: true, openWorldHint: true },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: true },
     }, ({ url, search, limit }) => run(async () => {
         const res = await client.map({ url, search, limit });
         return formatMap(res) + formatCredits(res);
@@ -138,7 +139,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             url: z.string().url(),
             max_pages: desc(z.number().int().min(1).max(25).optional(), "Default 5"),
         },
-        annotations: { readOnlyHint: true, openWorldHint: true },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: true },
     }, ({ url, max_pages }) => run(async () => {
         const res = await client.crawl({ url, maxPages: max_pages });
         return formatCrawl(res) + formatCredits(res);
@@ -153,7 +154,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             prompt: z.string().optional(),
             schema: desc(z.record(z.string(), z.any()).optional(), "JSON Schema for the output"),
         },
-        annotations: { readOnlyHint: true, openWorldHint: true },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: true },
     }, ({ url, prompt, schema }) => run(async () => {
         const res = await client.extract({ url, prompt, schema });
         return JSON.stringify(res.data ?? res, null, 2) + formatCredits(res);
@@ -168,7 +169,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             documents: z.array(z.string()).min(1).max(200),
             top_k: z.number().int().min(1).optional(),
         },
-        annotations: { readOnlyHint: true, openWorldHint: false },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: false },
     }, ({ query, documents, top_k }) => run(async () => {
         const res = await client.rerank({ query, documents, topK: top_k });
         const items = Array.isArray(res.results)
@@ -203,7 +204,8 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
             notify_email: desc(z.string().optional(), "Email address to notify when triggered, or 'true' to use the account email"),
             webhook_url: desc(z.string().url().optional(), "HTTPS webhook URL to POST when triggered"),
         },
-        annotations: { readOnlyHint: false, openWorldHint: true },
+        // Destructive: the `delete` action removes a monitor and its history.
+        annotations: { readOnlyHint: false, destructiveHint: true, openWorldHint: true },
     }, ({ action, id, url, urls, mode, max_pages, name, interval_minutes, watch_for, condition, notify_email, webhook_url, }) => run(async () => {
         const effectiveAction = action ?? (url || (urls && urls.length) ? "create" : "list");
         const requireId = (op) => {
@@ -284,7 +286,7 @@ export function createWebcrawlMcpServer(client, versionOrOpts = "0.1.0") {
         title: "Search categories",
         description: "List the search categories, subcategories and filters web_search supports.",
         inputSchema: {},
-        annotations: { readOnlyHint: true, openWorldHint: false },
+        annotations: { readOnlyHint: true, destructiveHint: false, openWorldHint: false },
     }, () => run(async () => JSON.stringify(await client.categories(), null, 2)));
     return server;
 }

package/package.json

@@ -1,11 +1,12 @@
 {
   "name": "@webcrawl/mcp",
-  "version": "1.3.8",
-  "description": "MCP server for Webcrawl Anything — zero-key local engine or webcrawl.app cloud (search, scrape, parse PDF/Office/YouTube, map, crawl, extract, rerank)",
+  "version": "1.3.10",
+  "mcpName": "io.github.brainfish-ai/webcrawl",
+  "description": "MCP server for Webcrawl Anything: search, scrape, map, crawl, parse and monitor the web from your agent. Runs locally with no API key, or on webcrawl.app cloud.",
   "type": "module",
   "license": "MIT",
   "author": "Brainfish AI <support@webcrawl.app> (https://webcrawl.app)",
-  "homepage": "https://webcrawl.app",
+  "homepage": "https://github.com/brainfish-ai/webcrawl-anything#readme",
   "repository": {
     "type": "git",
     "url": "git+https://github.com/brainfish-ai/webcrawl-anything.git",
@@ -52,7 +53,7 @@
   },
   "dependencies": {
     "@modelcontextprotocol/sdk": "^1.32.2",
-    "@webcrawl/client": "^1.3.8",
+    "@webcrawl/client": "^1.3.10",
     "zod": "^4.0.0"
   },
   "engines": {