@ever-works/jina-plugin 1.0.0 → 1.0.3

package/dist/index.js

@@ -0,0 +1,353 @@
+import { isIP } from 'net';
+
+// ../../plugin/dist/helpers/ssrf-guard.js
+function isSafeWebhookUrl(rawUrl) {
+  let url;
+  try {
+    url = new URL(rawUrl);
+  } catch {
+    return false;
+  }
+  if (url.protocol !== "https:" && url.protocol !== "http:") {
+    return false;
+  }
+  let host = url.hostname.toLowerCase();
+  if (host.startsWith("[") && host.endsWith("]")) {
+    host = host.slice(1, -1);
+  }
+  if (CLOUD_METADATA_HOSTNAMES.has(host)) {
+    return false;
+  }
+  const ipKind = isIP(host);
+  if (ipKind === 4 && isPrivateIPv4(host)) return false;
+  if (ipKind === 6 && isPrivateIPv6(host)) return false;
+  return true;
+}
+var CLOUD_METADATA_HOSTNAMES = /* @__PURE__ */ new Set([
+  // GCP
+  "metadata.google.internal",
+  "metadata.goog",
+  // AWS
+  "instance-data",
+  "instance-data.ec2.internal",
+  // Azure
+  "metadata.azure.internal"
+]);
+function isPrivateIPv4(ip) {
+  const parts = ip.split(".").map((p) => Number(p));
+  if (parts.length !== 4 || parts.some((n) => Number.isNaN(n))) return false;
+  const [a, b] = parts;
+  if (a === 10) return true;
+  if (a === 127) return true;
+  if (a === 0) return true;
+  if (a === 169 && b === 254) return true;
+  if (a === 172 && b >= 16 && b <= 31) return true;
+  if (a === 192 && b === 168) return true;
+  if (a === 192 && b === 0 && parts[2] === 0) return true;
+  if (a === 100 && b >= 64 && b <= 127) return true;
+  if (a >= 224) return true;
+  return false;
+}
+function isPrivateIPv6(ip) {
+  const lower = ip.toLowerCase();
+  if (lower === "::1" || lower === "::") return true;
+  if (lower.startsWith("fc") || lower.startsWith("fd")) return true;
+  const firstHextet = lower.split(":", 1)[0];
+  if (/^[0-9a-f]{1,4}$/.test(firstHextet)) {
+    const group0 = parseInt(firstHextet, 16);
+    if ((group0 & 65472) === 65152) return true;
+  }
+  if (lower.startsWith("::ffff:")) {
+    const v4 = lower.slice("::ffff:".length);
+    if (isIP(v4) === 4) return isPrivateIPv4(v4);
+  }
+  return false;
+}
+
+// src/jina.plugin.ts
+var JINA_READER_URL = "https://r.jina.ai/";
+var JINA_SEARCH_URL = "https://s.jina.ai/";
+var JinaReaderPlugin = class {
+  id = "jina";
+  name = "Jina AI";
+  version = "1.0.0";
+  category = "content-extractor";
+  capabilities = ["search", "content-extractor"];
+  providerName = "Jina";
+  settingsSchema = {
+    type: "object",
+    required: ["apiKey"],
+    properties: {
+      apiKey: {
+        type: "string",
+        title: "API Key",
+        description: "Your Jina API key. Get one at https://jina.ai",
+        "x-secret": true,
+        "x-envVar": "PLUGIN_JINA_API_KEY",
+        "x-scope": "user"
+      }
+    }
+  };
+  configurationMode = "hybrid";
+  context;
+  async search(options) {
+    const startTime = Date.now();
+    const apiKey = options.settings?.apiKey;
+    try {
+      const headers = {
+        Accept: "application/json",
+        "Content-Type": "application/json",
+        "X-Respond-With": "no-content",
+        Authorization: `Bearer ${apiKey}`
+      };
+      if (options.includeDomains && options.includeDomains.length > 0) {
+        headers["X-Site"] = options.includeDomains[0];
+      }
+      const body = {
+        q: options.query
+      };
+      if (options.limit) {
+        body.num = options.limit;
+      }
+      if (options.region) {
+        body.gl = options.region;
+      }
+      if (options.language) {
+        body.hl = options.language;
+      }
+      const response = await fetch(JINA_SEARCH_URL, {
+        method: "POST",
+        headers,
+        body: JSON.stringify(body),
+        signal: AbortSignal.timeout(3e4)
+      });
+      if (!response.ok) {
+        throw new Error(`Jina Search API returned ${response.status}: ${response.statusText}`);
+      }
+      const json = await response.json();
+      const results = (json.data || []).map((r, index) => {
+        let source;
+        try {
+          if (r.url) source = new URL(r.url).hostname;
+        } catch {
+        }
+        return {
+          title: r.title || "",
+          url: r.url || "",
+          snippet: r.description || "",
+          position: index + 1,
+          publishedDate: r.date || void 0,
+          source
+        };
+      });
+      return {
+        results,
+        query: options.query,
+        totalResults: results.length,
+        hasMore: false,
+        duration: Date.now() - startTime
+      };
+    } catch (error) {
+      this.context?.logger.error(`Jina search failed: ${error instanceof Error ? error.message : String(error)}`);
+      throw error;
+    }
+  }
+  async isAvailable() {
+    if (!this.context) return false;
+    const settings = await this.context.getSettings();
+    return Boolean(settings?.apiKey);
+  }
+  async validateConnection(settings) {
+    const apiKey = settings.apiKey;
+    if (!apiKey) {
+      return { success: false, message: "Jina AI API key is not configured." };
+    }
+    try {
+      const response = await fetch(JINA_SEARCH_URL, {
+        method: "POST",
+        headers: {
+          Accept: "application/json",
+          "Content-Type": "application/json",
+          "X-Respond-With": "no-content",
+          Authorization: `Bearer ${apiKey}`
+        },
+        body: JSON.stringify({ q: "test", num: 1 }),
+        signal: AbortSignal.timeout(15e3)
+      });
+      if (!response.ok) {
+        return {
+          success: false,
+          message: `Jina AI connection failed (${response.status}): ${response.statusText}`
+        };
+      }
+      return { success: true, message: "Jina AI connection verified." };
+    } catch (error) {
+      return {
+        success: false,
+        message: `Jina AI connection failed: ${error instanceof Error ? error.message : String(error)}`
+      };
+    }
+  }
+  async getRateLimitInfo() {
+    return {
+      remaining: -1,
+      limit: -1,
+      period: "minute"
+    };
+  }
+  async extract(options) {
+    const startTime = Date.now();
+    const { url, settings } = options;
+    const apiKey = settings?.apiKey;
+    if (!isSafeWebhookUrl(url)) {
+      return {
+        success: false,
+        url,
+        error: `URL host is not safe to fetch (SSRF guard blocked: ${url})`,
+        duration: Date.now() - startTime
+      };
+    }
+    try {
+      const headers = {
+        Accept: "application/json",
+        "X-Respond-With": "markdown",
+        Authorization: `Bearer ${apiKey}`
+      };
+      const response = await fetch(JINA_READER_URL, {
+        method: "POST",
+        headers: {
+          ...headers,
+          "Content-Type": "application/json"
+        },
+        body: JSON.stringify({ url }),
+        signal: AbortSignal.timeout(options.timeout || 3e4)
+      });
+      if (!response.ok) {
+        return {
+          success: false,
+          url,
+          error: `Jina API returned ${response.status}: ${response.statusText}`,
+          duration: Date.now() - startTime
+        };
+      }
+      const json = await response.json();
+      const data = json.data;
+      if (!data || !data.content) {
+        return {
+          success: false,
+          url,
+          error: "No content returned from Jina API",
+          duration: Date.now() - startTime
+        };
+      }
+      const wordCount = data.content.split(/\s+/).filter((w) => w.length > 0).length;
+      const images = data.images ? Object.entries(data.images).map(([alt, src]) => ({ src, alt })) : void 0;
+      const links = data.links ? Object.entries(data.links).map(([text, href]) => ({
+        href,
+        text,
+        isExternal: true
+      })) : void 0;
+      const metadata = {};
+      if (data.description) metadata.description = data.description;
+      if (data.publishedTime) metadata.publishedDate = data.publishedTime;
+      if (data.metadata?.lang) metadata.language = data.metadata.lang;
+      return {
+        success: true,
+        url,
+        finalUrl: data.url !== url ? data.url : void 0,
+        title: data.title,
+        content: data.content,
+        markdown: data.content,
+        images: options.includeImages !== false ? images : void 0,
+        links: options.includeLinks !== false ? links : void 0,
+        metadata: Object.keys(metadata).length > 0 ? metadata : void 0,
+        duration: Date.now() - startTime,
+        wordCount,
+        readingTime: Math.ceil(wordCount / 200)
+      };
+    } catch (error) {
+      return {
+        success: false,
+        url,
+        error: error instanceof Error ? error.message : String(error),
+        duration: Date.now() - startTime
+      };
+    }
+  }
+  async extractBatch(urls, options) {
+    const batchSize = 5;
+    const results = [];
+    for (let i = 0; i < urls.length; i += batchSize) {
+      const batch = urls.slice(i, i + batchSize);
+      const batchResults = await Promise.all(batch.map((url) => this.extract({ url, ...options })));
+      results.push(...batchResults);
+      if (i + batchSize < urls.length) {
+        await new Promise((resolve) => setTimeout(resolve, 100));
+      }
+    }
+    return results;
+  }
+  async canExtract(url) {
+    return isSafeWebhookUrl(url);
+  }
+  getSupportedFormats() {
+    return ["text", "markdown"];
+  }
+  async onLoad(context) {
+    this.context = context;
+    context.logger.log("Jina AI Reader Plugin loaded");
+  }
+  async onUnload() {
+    this.context = void 0;
+  }
+  async healthCheck() {
+    return {
+      status: "healthy",
+      message: "Jina AI Reader plugin is ready",
+      checkedAt: Date.now()
+    };
+  }
+  getManifest() {
+    return {
+      id: this.id,
+      name: this.name,
+      version: this.version,
+      description: "Web search and content extraction using the Jina AI APIs",
+      category: this.category,
+      capabilities: [...this.capabilities],
+      author: { name: "Ever Works Team" },
+      license: "AGPL-3.0",
+      builtIn: true,
+      systemPlugin: false,
+      autoEnable: false,
+      homepage: "https://jina.ai",
+      icon: {
+        type: "url",
+        value: "https://jina.ai/icons/favicon-128x128.png"
+      },
+      readme: [
+        "## What does Jina AI do?",
+        "",
+        "Jina AI provides web search with LLM-optimized results and content extraction that converts any web page into clean markdown.",
+        "",
+        "## Why use it?",
+        "",
+        "- **Web search** \u2014 search the web and get results with content already extracted",
+        "- **Content extraction** \u2014 converts pages to clean markdown, strips ads and navigation",
+        "- **Domain filtering** \u2014 restrict search to specific domains",
+        "",
+        "## How it works in Ever Works",
+        "",
+        "During work generation, Jina finds relevant information about each item via search and extracts clean content from web pages for enriching descriptions.",
+        "",
+        "## Getting started",
+        "",
+        "1. Get an API key at [jina.ai](https://jina.ai)",
+        "2. Enter the key in the **API Key** field below",
+        "3. Enable the plugin"
+      ].join("\n")
+    };
+  }
+};
+
+export { JinaReaderPlugin, JinaReaderPlugin as default };

package/package.json

@@ -1,75 +1,88 @@
 {
-  "name": "@ever-works/jina-plugin",
-  "version": "1.0.0",
-  "description": "Jina AI Plugin - Web search and content extraction using Jina AI APIs",
-  "author": {
-    "name": "Ever Co. LTD",
-    "email": "ever@ever.co",
-    "url": "https://ever.co"
-  },
-  "license": "AGPL-3.0",
-  "private": false,
-  "type": "module",
-  "main": "./dist/index.cjs",
-  "module": "./dist/index.js",
-  "types": "./dist/index.d.ts",
-  "exports": {
-    ".": {
-      "types": "./dist/index.d.ts",
-      "import": "./dist/index.js",
-      "require": "./dist/index.cjs"
-    }
-  },
-  "peerDependencies": {
-    "@ever-works/plugin": "1.1.0"
-  },
-  "devDependencies": {
-    "@types/node": "^22.0.0",
-    "tsup": "^8.4.0",
-    "typescript": "^5.7.3",
-    "vitest": "^4.1.0",
-    "@ever-works/plugin": "1.1.0"
-  },
-  "everworks": {
-    "plugin": {
-      "id": "jina",
-      "name": "Jina AI",
-      "version": "1.0.0",
-      "category": "content-extractor",
-      "capabilities": [
-        "search",
-        "content-extractor"
-      ],
-      "description": "Web search and content extraction using Jina AI APIs. Search via s.jina.ai and extract content via r.jina.ai, no SDK needed.",
-      "author": {
-        "name": "Ever Works Team"
-      },
-      "license": "AGPL-3.0",
-      "autoEnable": false,
-      "systemPlugin": false,
-      "builtIn": true,
-      "envVars": [
-        {
-          "name": "PLUGIN_JINA_API_KEY",
-          "required": true,
-          "secret": true,
-          "description": "Jina API key"
-        }
-      ],
-      "distribution": "registry"
-    }
-  },
-  "publishConfig": {
-    "access": "restricted",
-    "registry": "https://registry.npmjs.org"
-  },
-  "scripts": {
-    "build": "tsc --noEmit && tsup",
-    "dev": "tsup --watch",
-    "type-check": "tsc --noEmit",
-    "clean": "rm -rf dist",
-    "test": "vitest run --passWithNoTests",
-    "test:watch": "vitest",
-    "test:coverage": "vitest run --coverage"
-  }
-}
+	"name": "@ever-works/jina-plugin",
+	"version": "1.0.3",
+	"description": "Jina AI Plugin - Web search and content extraction using Jina AI APIs",
+	"author": {
+		"name": "Ever Co. LTD",
+		"email": "ever@ever.co",
+		"url": "https://ever.co"
+	},
+	"license": "AGPL-3.0",
+	"private": false,
+	"type": "module",
+	"main": "./dist/index.cjs",
+	"module": "./dist/index.js",
+	"types": "./dist/index.d.ts",
+	"exports": {
+		".": {
+			"types": "./dist/index.d.ts",
+			"import": "./dist/index.js",
+			"require": "./dist/index.cjs"
+		}
+	},
+	"files": [
+		"dist"
+	],
+	"scripts": {
+		"build": "tsc --noEmit && tsup",
+		"dev": "tsup --watch",
+		"type-check": "tsc --noEmit",
+		"clean": "rm -rf dist",
+		"test": "vitest run --passWithNoTests",
+		"test:watch": "vitest",
+		"test:coverage": "vitest run --coverage"
+	},
+	"peerDependencies": {
+		"@ever-works/plugin": "^1.2.5"
+	},
+	"devDependencies": {
+		"@types/node": "^26.6.4",
+		"@ever-works/plugin": "^1.2.5",
+		"tsup": "^8.4.0",
+		"typescript": "^5.7.3",
+		"vitest": "^5.0.5"
+	},
+	"everworks": {
+		"plugin": {
+			"id": "jina",
+			"name": "Jina AI",
+			"version": "1.0.0",
+			"category": "content-extractor",
+			"capabilities": [
+				"search",
+				"content-extractor"
+			],
+			"description": "Web search and content extraction using Jina AI APIs. Search via s.jina.ai and extract content via r.jina.ai, no SDK needed.",
+			"author": {
+				"name": "Ever Works Team"
+			},
+			"license": "AGPL-3.0",
+			"autoEnable": false,
+			"systemPlugin": false,
+			"builtIn": true,
+			"envVars": [
+				{
+					"name": "PLUGIN_JINA_API_KEY",
+					"required": true,
+					"secret": true,
+					"description": "Jina API key"
+				}
+			],
+			"distribution": "registry"
+		}
+	},
+	"repository": {
+		"type": "git",
+		"url": "git+https://github.com/ever-works/ever-works.git",
+		"directory": "packages/plugins/jina"
+	},
+	"homepage": "https://github.com/ever-works/ever-works/tree/main/packages/plugins/jina#readme",
+	"bugs": {
+		"url": "https://github.com/ever-works/ever-works/issues"
+	},
+	"everworksRelease": {
+		"contentHash": "sha256:38644b829f1b71573e5aca4b3bd20edb1809e2886bad28b2bf80f6315199e61d",
+		"commit": "19905c70cf05be3bc523db030597046067cd322a",
+		"source": "https://github.com/ever-works/ever-works/tree/19905c70cf05be3bc523db030597046067cd322a/packages/plugins/jina"
+	}
+}