@metalift/mcp - npm Package Compare versions

Comparing version 1.0.5 to 1.0.6

dist/client.js

@@ -22,2 +22,31 @@ const B2B_ATTESTATION_HEADER = "X-B2B-Attestation";
 }
+function formatApiError(body, status) {
+    if (typeof body === "object" && body !== null) {
+        const record = body;
+        const detail = record.detail;
+        if (typeof detail === "string" && detail) {
+            return detail;
+        }
+        if (typeof detail === "object" && detail !== null) {
+            const detailRecord = detail;
+            const message = detailRecord.message;
+            const code = detailRecord.code;
+            if (typeof message === "string" && message) {
+                return typeof code === "string" ? `${message} (${code})` : message;
+            }
+        }
+        const error = record.error;
+        if (typeof error === "string" && error) {
+            return error;
+        }
+    }
+    if (typeof body === "string" && body.trim()) {
+        const trimmed = body.trim();
+        if (trimmed.startsWith("<!DOCTYPE") || trimmed.startsWith("<html")) {
+            return `Request failed: ${status} (received HTML instead of JSON — check METALIFT_API_URL points to the scrape API, e.g. http://localhost:8080)`;
+        }
+        return trimmed.slice(0, 500);
+    }
+    return `Request failed: ${status}`;
+}
 export class MetaliftClient {
@@ -46,14 +75,12 @@ apiUrl;
         });
-        const body = (await response.json());
+        const rawBody = await response.text();
+        let body;
+        try {
+            body = rawBody ? JSON.parse(rawBody) : {};
+        }
+        catch {
+            throw new Error(formatApiError(rawBody, response.status));
+        }
         if (!response.ok) {
-            const detail = typeof body === "object" && body !== null && "detail" in body
-                ? body.detail
-                : undefined;
-            const error = typeof body === "object" && body !== null && "error" in body
-                ? body.error
-                : undefined;
-            const message = (typeof detail === "string" && detail) ||
-                (typeof error === "string" && error) ||
-                `Request failed: ${response.status}`;
-            throw new Error(message);
+            throw new Error(formatApiError(body, response.status));
         }
@@ -60,0 +87,0 @@ return withBilling(body, billingFromResponse(response));

dist/index.js

@@ -15,3 +15,3 @@ #!/usr/bin/env node
     title: "Scrape URL",
-    description: `Scrape a single URL into markdown, HTML, or text for LLM context. Default: fast direct static article extraction (strategy=article, render=static, proxy=direct, 10s timeout). For WAF, SPA, retail, or JS-heavy pages pass strategy=auto or a specific strategy (spa, cloudflare, retail). Response includes credits_charged based on actual usage (static=1, JS=5, premium=10+). ${COMPLIANCE_NOTICE}`,
+    description: `Scrape a single URL into markdown, HTML, or text for LLM context. Default: fast direct static article extraction (strategy=article, render=static, proxy=direct, 10s timeout). For full page HTML on static sites use strategy=download with formats=["html"] (1 credit, all tiers). strategy=raw and full-page HTML without download require Enterprise tier. For WAF, SPA, retail, or JS-heavy pages pass strategy=auto or a specific strategy (spa, cloudflare, retail). Response includes credits_charged based on actual usage (static=1, JS=5, premium=10+). ${COMPLIANCE_NOTICE}`,
     inputSchema: {
@@ -29,3 +29,3 @@ url: z.string().url(),
             .optional()
-            .describe("Scrape strategy: auto, article, spa, cloudflare, authenticated, listing, retail, jsonld, raw, or comma-separated chain. See /v1/strategies for protection levels and credit estimates."),
+            .describe("Scrape strategy: auto, article, spa, cloudflare, authenticated, listing, retail, jsonld, download, raw, or comma-separated chain. Use download for full static HTML. See /v1/strategies for protection levels and credit estimates."),
         cookies: z.record(z.string()).optional(),
@@ -32,0 +32,0 @@ cookie_header: z

dist/scrape-args.js

 /** Default timeout for the fast direct static markdown path. */
 export const FAST_SCRAPE_TIMEOUT_MS = 10_000;
+function wantsFullPageHtml(args) {
+    return args.formats?.includes("html") === true && args.only_main_content === false;
+}
+function shouldAutoDownload(args) {
+    if (!wantsFullPageHtml(args)) {
+        return false;
+    }
+    const primary = (args.strategy ?? "auto").split(",")[0]?.trim().toLowerCase();
+    if (primary === "raw" || primary === "download") {
+        return false;
+    }
+    if (primary && primary !== "auto" && primary !== "article") {
+        return false;
+    }
+    return true;
+}
 function isMarkdownOnlyFormats(formats) {
@@ -11,2 +27,11 @@ return formats === undefined || (formats.length === 1 && formats[0] === "markdown");
 export function normalizeScrapeArgs(args) {
+    if (shouldAutoDownload(args)) {
+        return {
+            ...args,
+            strategy: "download",
+            render: args.render ?? "static",
+            proxy: args.proxy ?? "direct",
+            timeout_ms: args.timeout_ms ?? FAST_SCRAPE_TIMEOUT_MS,
+        };
+    }
     const hasAdvancedOpts = args.strategy !== undefined ||
@@ -27,6 +52,5 @@ args.render !== undefined ||
         strategy: "article",
-        render: "static",
-        proxy: "direct",
+        render: "auto",
         timeout_ms: args.timeout_ms ?? FAST_SCRAPE_TIMEOUT_MS,
     };
 }

dist/scrape-args.test.js

@@ -10,4 +10,3 @@ import assert from "node:assert/strict";
         strategy: "article",
-        render: "static",
-        proxy: "direct",
+        render: "auto",
         timeout_ms: FAST_SCRAPE_TIMEOUT_MS,
@@ -64,2 +63,24 @@ });
 });
+test("normalizeScrapeArgs auto-selects download for full-page HTML", () => {
+    const args = normalizeScrapeArgs({
+        url: BASE_URL,
+        formats: ["html"],
+        only_main_content: false,
+    });
+    assert.equal(args.strategy, "download");
+    assert.equal(args.render, "static");
+    assert.equal(args.proxy, "direct");
+    assert.equal(args.timeout_ms, FAST_SCRAPE_TIMEOUT_MS);
+});
+test("normalizeScrapeArgs preserves explicit cloudflare for full-page HTML", () => {
+    const args = normalizeScrapeArgs({
+        url: BASE_URL,
+        formats: ["html"],
+        only_main_content: false,
+        strategy: "cloudflare",
+        render: "dynamic",
+    });
+    assert.equal(args.strategy, "cloudflare");
+    assert.equal(args.render, "dynamic");
+});
 test("normalizeScrapeArgs preserves non-markdown formats", () => {
@@ -77,4 +98,3 @@ const args = normalizeScrapeArgs({ url: BASE_URL, formats: ["html"] });
     assert.equal(args.strategy, "article");
-    assert.equal(args.render, "static");
-    assert.equal(args.proxy, "direct");
+    assert.equal(args.render, "auto");
 });

package.json

 {
   "name": "@metalift/mcp",
   "mcpName": "io.github.MetaLift-AI/metalift",
-  "version": "1.0.5",
+  "version": "1.0.6",
   "description": "Metalift MCP server for AI agents",
@@ -6,0 +6,0 @@ "license": "SEE LICENSE IN LICENSE",

New alerts

Unidentified License

License

Something that seems like a license was found, but its contents could not be matched with a known license.

Fixed alerts

AI-detected potential code anomaly

Supply chain risk

AI has identified unusual behaviors that may pose a security risk.

Unidentified License

License

Something that seems like a license was found, but its contents could not be matched with a known license.

Improved metrics

Total package byte prevSize

30718

9.2%

Lines of code

616

13.03%

Number of low supply chain risk alerts

3

-25%

No dependency changes