agentkit-cleanweb 0.0.0-stage → 1.0.1

package/dist/provider.js

@@ -0,0 +1,99 @@
+import { CleanWebContentSchema, CleanPdfSchema, CleanYoutubeSchema, OracleGroundingSchema, ManageVaultSchema, } from "./types.js";
+import { cleanWebContent, cleanPdfResearch, cleanYoutubeTranscript, oracleGrounding, manageX402Vault, } from "./actions.js";
+import { BASE_CHAIN_ID, BASE_SEPOLIA_CHAIN_ID } from "./constants.js";
+/**
+ * Coinbase AgentKit CleanWeb Action Provider
+ * Drop-in Action Provider supplying 87% token-compressed web scraping,
+ * research PDF analysis, YouTube Gemini Flash intelligence, and Base EIP-712 oracle attestations.
+ */
+export class CleanWebActionProvider {
+    name = "cleanweb";
+    description = "87% token noise reduction web scraper, academic PDF analyzer, YouTube Gemini video intelligence, and Base Mainnet EIP-712 signed oracle grounding.";
+    config;
+    constructor(config = {}) {
+        this.config = config;
+    }
+    /**
+     * Verifies if the provider supports the given network.
+     * Native support for Base (8453), Base Sepolia (84532), and all EVM chains.
+     */
+    supportsNetwork(network) {
+        if (!network || !network.chainId)
+            return true;
+        const cid = Number(network.chainId);
+        return cid === BASE_CHAIN_ID || cid === BASE_SEPOLIA_CHAIN_ID || cid === 137 || cid === 42161;
+    }
+    /**
+     * Returns all 5 Action instances for Coinbase AgentKit runtime registration.
+     */
+    getActions(walletProvider) {
+        const opts = {
+            gatewayUrl: this.config.gatewayUrl,
+            vaultKey: this.config.vaultKey,
+            agentNonce: this.config.agentNonce,
+            timeoutMs: this.config.timeoutMs,
+        };
+        return [
+            {
+                name: "cleanweb_clean_web_content",
+                description: "Scrapes any target webpage URL and compresses noise/ads into 87% token-reduced clean Markdown for LLMs.",
+                schema: CleanWebContentSchema,
+                invoke: async (args) => {
+                    const parsed = CleanWebContentSchema.parse(args);
+                    return await cleanWebContent(parsed, opts);
+                },
+            },
+            {
+                name: "cleanweb_clean_pdf_research",
+                description: "Extracts structured academic text, executive findings, and sections from online PDF whitepapers and reports.",
+                schema: CleanPdfSchema,
+                invoke: async (args) => {
+                    const parsed = CleanPdfSchema.parse(args);
+                    return await cleanPdfResearch(parsed, opts);
+                },
+            },
+            {
+                name: "cleanweb_clean_youtube_transcript",
+                description: "Extracts high-precision subtitles and Google Gemini 3.8 Flash summary from YouTube videos without downloading media.",
+                schema: CleanYoutubeSchema,
+                invoke: async (args) => {
+                    const parsed = CleanYoutubeSchema.parse(args);
+                    return await cleanYoutubeTranscript(parsed, opts);
+                },
+            },
+            {
+                name: "cleanweb_oracle_grounding",
+                description: "Discovers factual ground truth from the web and cryptographically signs an on-chain verifiable EIP-712 attestation for Base (8453).",
+                schema: OracleGroundingSchema,
+                invoke: async (args) => {
+                    const parsed = OracleGroundingSchema.parse(args);
+                    return await oracleGrounding(parsed, opts);
+                },
+            },
+            {
+                name: "cleanweb_manage_x402_vault",
+                description: "Checks prefunded agent USDC vault balance, retrieves Base USDC deposit address, or inspects VIP pass entitlements.",
+                schema: ManageVaultSchema,
+                invoke: async (args) => {
+                    const parsed = ManageVaultSchema.parse(args);
+                    let addr = parsed.agentAddress;
+                    if (!addr && walletProvider && typeof walletProvider.getAddress === "function") {
+                        try {
+                            addr = await walletProvider.getAddress();
+                        }
+                        catch {
+                            // fallback
+                        }
+                    }
+                    return await manageX402Vault({ ...parsed, agentAddress: addr }, opts);
+                },
+            },
+        ];
+    }
+}
+/**
+ * Factory helper for clean instantiations
+ */
+export function cleanWebActionProvider(config = {}) {
+    return new CleanWebActionProvider(config);
+}

package/dist/types.d.ts

@@ -0,0 +1,137 @@
+import { z } from "zod";
+/**
+ * 1. Clean Web Content Schema
+ */
+export declare const CleanWebContentSchema: z.ZodObject<{
+    url: z.ZodString;
+    maxLength: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
+    includeLinks: z.ZodDefault<z.ZodOptional<z.ZodBoolean>>;
+}, "strip", z.ZodTypeAny, {
+    url: string;
+    maxLength: number;
+    includeLinks: boolean;
+}, {
+    url: string;
+    maxLength?: number | undefined;
+    includeLinks?: boolean | undefined;
+}>;
+export type CleanWebContentParams = z.infer<typeof CleanWebContentSchema>;
+/**
+ * 2. Clean Research PDF Schema
+ */
+export declare const CleanPdfSchema: z.ZodObject<{
+    pdfUrl: z.ZodString;
+    maxPages: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
+}, "strip", z.ZodTypeAny, {
+    pdfUrl: string;
+    maxPages: number;
+}, {
+    pdfUrl: string;
+    maxPages?: number | undefined;
+}>;
+export type CleanPdfParams = z.infer<typeof CleanPdfSchema>;
+/**
+ * 3. Clean YouTube Transcript Schema
+ */
+export declare const CleanYoutubeSchema: z.ZodObject<{
+    videoUrl: z.ZodString;
+    includeSummary: z.ZodDefault<z.ZodOptional<z.ZodBoolean>>;
+}, "strip", z.ZodTypeAny, {
+    videoUrl: string;
+    includeSummary: boolean;
+}, {
+    videoUrl: string;
+    includeSummary?: boolean | undefined;
+}>;
+export type CleanYoutubeParams = z.infer<typeof CleanYoutubeSchema>;
+/**
+ * 4. Oracle Grounding & EIP-712 Attestation Schema
+ */
+export declare const OracleGroundingSchema: z.ZodObject<{
+    query: z.ZodString;
+    targetUrl: z.ZodOptional<z.ZodString>;
+    chainId: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
+}, "strip", z.ZodTypeAny, {
+    query: string;
+    chainId: number;
+    targetUrl?: string | undefined;
+}, {
+    query: string;
+    targetUrl?: string | undefined;
+    chainId?: number | undefined;
+}>;
+export type OracleGroundingParams = z.infer<typeof OracleGroundingSchema>;
+/**
+ * 5. Manage x402 Agent Vault Schema
+ */
+export declare const ManageVaultSchema: z.ZodObject<{
+    action: z.ZodEnum<["balance", "deposit_info", "pass_status"]>;
+    agentAddress: z.ZodOptional<z.ZodString>;
+}, "strip", z.ZodTypeAny, {
+    action: "balance" | "deposit_info" | "pass_status";
+    agentAddress?: string | undefined;
+}, {
+    action: "balance" | "deposit_info" | "pass_status";
+    agentAddress?: string | undefined;
+}>;
+export type ManageVaultParams = z.infer<typeof ManageVaultSchema>;
+/**
+ * CleanWeb Universal Action Result Interfaces
+ */
+export interface CleanWebContentResult {
+    success: boolean;
+    url: string;
+    title: string;
+    cleanMarkdown: string;
+    tokenReductionRate: string;
+    sourceTokensEstimate: number;
+    compressedTokensEstimate: number;
+    latencyMs: number;
+}
+export interface CleanPdfResult {
+    success: boolean;
+    pdfUrl: string;
+    title: string;
+    pageCount: number;
+    wordCount: number;
+    text: string;
+    latencyMs: number;
+}
+export interface CleanYoutubeResult {
+    success: boolean;
+    videoId: string;
+    title: string;
+    channel: string;
+    summary?: string;
+    transcriptSample: string;
+    fullTranscriptLength: number;
+    latencyMs: number;
+}
+export interface OracleGroundingResult {
+    success: boolean;
+    query: string;
+    groundedFacts: string[];
+    summary: string;
+    verdict: "VERIFIED" | "CONTRADICTED" | "UNVERIFIED";
+    confidence: number;
+    attestation?: {
+        chainId: number;
+        signerAddress: string;
+        merkleRoot: string;
+        signature: string;
+        eip712Domain: Record<string, unknown>;
+    };
+    latencyMs: number;
+}
+export interface ManageVaultResult {
+    success: boolean;
+    action: string;
+    agentAddress?: string;
+    balanceUsdc?: number;
+    totalConsumedUsdc?: number;
+    depositAddress?: string;
+    supportedChains?: string[];
+    passTier?: string;
+    passActive?: boolean;
+}
+//# sourceMappingURL=types.d.ts.map

package/dist/types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAExB;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;EAiB4C,CAAC;AAE/E,MAAM,MAAM,qBAAqB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,qBAAqB,CAAC,CAAC;AAE1E;;GAEG;AACH,eAAO,MAAM,cAAc;;;;;;;;;EAa2E,CAAC;AAEvG,MAAM,MAAM,cAAc,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,cAAc,CAAC,CAAC;AAE5D;;GAEG;AACH,eAAO,MAAM,kBAAkB;;;;;;;;;EAUoE,CAAC;AAEpG,MAAM,MAAM,kBAAkB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,kBAAkB,CAAC,CAAC;AAEpE;;GAEG;AACH,eAAO,MAAM,qBAAqB;;;;;;;;;;;;EAgBiF,CAAC;AAEpH,MAAM,MAAM,qBAAqB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,qBAAqB,CAAC,CAAC;AAE1E;;GAEG;AACH,eAAO,MAAM,iBAAiB;;;;;;;;;EAYsE,CAAC;AAErG,MAAM,MAAM,iBAAiB,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,iBAAiB,CAAC,CAAC;AAElE;;GAEG;AACH,MAAM,WAAW,qBAAqB;IACpC,OAAO,EAAE,OAAO,CAAC;IACjB,GAAG,EAAE,MAAM,CAAC;IACZ,KAAK,EAAE,MAAM,CAAC;IACd,aAAa,EAAE,MAAM,CAAC;IACtB,kBAAkB,EAAE,MAAM,CAAC;IAC3B,oBAAoB,EAAE,MAAM,CAAC;IAC7B,wBAAwB,EAAE,MAAM,CAAC;IACjC,SAAS,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,WAAW,cAAc;IAC7B,OAAO,EAAE,OAAO,CAAC;IACjB,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;IACd,SAAS,EAAE,MAAM,CAAC;IAClB,SAAS,EAAE,MAAM,CAAC;IAClB,IAAI,EAAE,MAAM,CAAC;IACb,SAAS,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,WAAW,kBAAkB;IACjC,OAAO,EAAE,OAAO,CAAC;IACjB,OAAO,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,MAAM,CAAC;IACd,OAAO,EAAE,MAAM,CAAC;IAChB,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,gBAAgB,EAAE,MAAM,CAAC;IACzB,oBAAoB,EAAE,MAAM,CAAC;IAC7B,SAAS,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,WAAW,qBAAqB;IACpC,OAAO,EAAE,OAAO,CAAC;IACjB,KAAK,EAAE,MAAM,CAAC;IACd,aAAa,EAAE,MAAM,EAAE,CAAC;IACxB,OAAO,EAAE,MAAM,CAAC;IAChB,OAAO,EAAE,UAAU,GAAG,cAAc,GAAG,YAAY,CAAC;IACpD,UAAU,EAAE,MAAM,CAAC;IACnB,WAAW,CAAC,EAAE;QACZ,OAAO,EAAE,MAAM,CAAC;QAChB,aAAa,EAAE,MAAM,CAAC;QACtB,UAAU,EAAE,MAAM,CAAC;QACnB,SAAS,EAAE,MAAM,CAAC;QAClB,YAAY,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;KACvC,CAAC;IACF,SAAS,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,WAAW,iBAAiB;IAChC,OAAO,EAAE,OAAO,CAAC;IACjB,MAAM,EAAE,MAAM,CAAC;IACf,YAAY,CAAC,EAAE,MAAM,CAAC;IACtB,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,iBAAiB,CAAC,EAAE,MAAM,CAAC;IAC3B,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,eAAe,CAAC,EAAE,MAAM,EAAE,CAAC;IAC3B,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,UAAU,CAAC,EAAE,OAAO,CAAC;CACtB"}

package/dist/types.js

@@ -0,0 +1,89 @@
+import { z } from "zod";
+/**
+ * 1. Clean Web Content Schema
+ */
+export const CleanWebContentSchema = z
+    .object({
+    url: z.string().url().describe("The HTTP/HTTPS webpage URL to scrape and compress into clean Markdown"),
+    maxLength: z
+        .number()
+        .int()
+        .positive()
+        .optional()
+        .default(50000)
+        .describe("Maximum character length for the returned markdown body (default 50,000)"),
+    includeLinks: z
+        .boolean()
+        .optional()
+        .default(true)
+        .describe("Whether to preserve relevant markdown links in the cleaned text"),
+})
+    .strip()
+    .describe("Parameters for scraping and reducing webpage token noise by 87%");
+/**
+ * 2. Clean Research PDF Schema
+ */
+export const CleanPdfSchema = z
+    .object({
+    pdfUrl: z.string().url().describe("The direct URL of the PDF whitepaper, financial filing, or arXiv research report"),
+    maxPages: z
+        .number()
+        .int()
+        .min(1)
+        .max(100)
+        .optional()
+        .default(50)
+        .describe("Maximum number of pages to parse (1-100, default 50)"),
+})
+    .strip()
+    .describe("Parameters for extracting clean academic and financial text from online PDF whitepapers");
+/**
+ * 3. Clean YouTube Transcript Schema
+ */
+export const CleanYoutubeSchema = z
+    .object({
+    videoUrl: z.string().describe("YouTube video URL or 11-character video ID"),
+    includeSummary: z
+        .boolean()
+        .optional()
+        .default(true)
+        .describe("Whether to generate Gemini 3.8 Flash bullet-point executive summary"),
+})
+    .strip()
+    .describe("Parameters for extracting timestamped subtitles and AI summaries from YouTube videos");
+/**
+ * 4. Oracle Grounding & EIP-712 Attestation Schema
+ */
+export const OracleGroundingSchema = z
+    .object({
+    query: z.string().min(3).describe("The factual proposition or topic to verify (e.g. 'Base TVL surpasses 4B' or 'CRMA compliance')"),
+    targetUrl: z
+        .string()
+        .url()
+        .optional()
+        .describe("Optional target webpage URL to ground the facts against"),
+    chainId: z
+        .number()
+        .int()
+        .optional()
+        .default(8453)
+        .describe("Target blockchain network ID for EIP-712 signature (Base Mainnet: 8453, Base Sepolia: 84532)"),
+})
+    .strip()
+    .describe("Parameters for web ground-truth discovery with on-chain verifiable EIP-712 cryptographic attestation");
+/**
+ * 5. Manage x402 Agent Vault Schema
+ */
+export const ManageVaultSchema = z
+    .object({
+    action: z
+        .enum(["balance", "deposit_info", "pass_status"])
+        .describe("Action to perform: 'balance' (check prefunded USDC balance), 'deposit_info' (get deposit address & chain parameters), or 'pass_status' (check agent VIP pass)"),
+    agentAddress: z
+        .string()
+        .regex(/^0x[a-fA-F0-9]{40}$/, "Must be a valid EVM address (0x...)")
+        .optional()
+        .describe("The agent's EVM wallet address on Base Mainnet"),
+})
+    .strip()
+    .describe("Parameters for managing gasless sub-millisecond x402 USDC micropayment session vaults");

package/package.json

@@ -1,6 +1,57 @@
 {
   "name": "agentkit-cleanweb",
-  "version": "0.0.0-stage",
-  "stub": true,
-  "description": "Temporary package placeholder for staged publishing"
-}
+  "version": "1.0.1",
+  "description": "Official Coinbase AgentKit Action Provider for CleanWeb 87% Token Reduction, Research PDF Parser, YouTube Gemini Intelligence, and Base Mainnet EIP-712 Oracle Grounding",
+  "main": "dist/index.js",
+  "module": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "type": "module",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/nohosa001-pixel/x402-cleanweb-agent.git",
+    "directory": "packages/agentkit-cleanweb"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
+  "scripts": {
+    "build": "tsc",
+    "test": "node --test tests/test_action_provider.test.mjs"
+  },
+  "keywords": [
+    "coinbase",
+    "agentkit",
+    "base",
+    "x402",
+    "ai-agent",
+    "web-scraping",
+    "token-reduction",
+    "ground-truth",
+    "eip712",
+    "oracle",
+    "gemini"
+  ],
+  "author": "CleanWeb & A.GRID Team",
+  "license": "MIT",
+  "peerDependencies": {
+    "@coinbase/agentkit": ">=0.1.0"
+  },
+  "peerDependenciesMeta": {
+    "@coinbase/agentkit": {
+      "optional": true
+    }
+  },
+  "dependencies": {
+    "zod": "^3.23.8"
+  },
+  "devDependencies": {
+    "@types/node": "^22.13.0",
+    "typescript": "^5.8.2",
+    "zod": "^3.23.8"
+  },
+  "files": [
+    "dist",
+    "src",
+    "README.md"
+  ]
+}