feat(fetch): added optional Kagi summarizer toggle to fetch configuration

- Added `fetch.useKagiSummarizer` configuration setting to toggle Kagi Universal Summarizer usage in fetch tool. - Updated fetch tool to conditionally apply Kagi summarization based on configuration setting. - Added comprehensive test coverage for Kagi summarizer toggle behavior with mocked dependencies.
2026-03-04 02:34:47 +01:00
parent 894bfe5a71
commit d8384a488e
4 changed files with 129 additions and 11 deletions
@@ -1,6 +1,9 @@
 # Changelog

 ## [Unreleased]
+### Added
+- Added `fetch.useKagiSummarizer` setting to toggle Kagi Universal Summarizer usage in the fetch tool.
+
 ### Fixed

 - Fixed incorrect message history reference in session title generation that could cause missing or stale titles on first message
@@ -490,6 +490,15 @@ export const SETTINGS_SCHEMA = {
 		default: true,
 		ui: { tab: "tools", label: "Enable Fetch", description: "Enable the fetch tool for URL fetching" },
 	},
+	"fetch.useKagiSummarizer": {
+		type: "boolean",
+		default: true,
+		ui: {
+			tab: "tools",
+			label: "Use Kagi in Fetch",
+			description: "Use Kagi Universal Summarizer when rendering HTML in fetch",
+		},
+	},
 	"web_search.enabled": {
 		type: "boolean",
 		default: true,
@@ -429,6 +429,7 @@ async function renderHtmlToText(
 	url: string,
 	html: string,
 	timeout: number,
+	useKagiSummarizer: boolean,
 	userSignal?: AbortSignal,
 ): Promise<{ content: string; ok: boolean; method: string }> {
 	const signal = ptree.combineSignals(userSignal, timeout * 1000);
@@ -440,15 +441,17 @@ async function renderHtmlToText(
 		signal,
 	};

-	// Try Kagi Universal Summarizer first (if KAGI_API_KEY is configured)
-	try {
-		const kagiSummary = await summarizeUrlWithKagi(url, { signal });
-		if (kagiSummary && kagiSummary.length > 100 && !isLowQualityOutput(kagiSummary)) {
-			return { content: kagiSummary, ok: true, method: "kagi" };
+	// Try Kagi Universal Summarizer first (if enabled and KAGI_API_KEY is configured)
+	if (useKagiSummarizer) {
+		try {
+			const kagiSummary = await summarizeUrlWithKagi(url, { signal });
+			if (kagiSummary && kagiSummary.length > 100 && !isLowQualityOutput(kagiSummary)) {
+				return { content: kagiSummary, ok: true, method: "kagi" };
+			}
+		} catch {
+			// Kagi failed, continue to next method
+			signal?.throwIfAborted();
 		}
-	} catch {
-		// Kagi failed, continue to next method
-		signal?.throwIfAborted();
 	}

 	// Try jina next (reader API)
@@ -564,7 +567,13 @@ async function handleSpecialUrls(url: string, timeout: number, signal?: AbortSig
 /**
 * Main render function implementing the full pipeline
 */
-async function renderUrl(url: string, timeout: number, raw: boolean, signal?: AbortSignal): Promise<RenderResult> {
+async function renderUrl(
+	url: string,
+	timeout: number,
+	raw: boolean,
+	useKagiSummarizer: boolean,
+	signal?: AbortSignal,
+): Promise<RenderResult> {
 	const notes: string[] = [];
 	const fetchedAt = new Date().toISOString();
 	if (signal?.aborted) {
@@ -803,7 +812,7 @@ async function renderUrl(url: string, timeout: number, raw: boolean, signal?: Ab
 		}

 		// Step 6: Render HTML with lynx or html2text
-		const htmlResult = await renderHtmlToText(finalUrl, rawContent, timeout, signal);
+		const htmlResult = await renderHtmlToText(finalUrl, rawContent, timeout, useKagiSummarizer, signal);
 		if (!htmlResult.ok) {
 			notes.push("html rendering failed (lynx/html2text unavailable)");
 			const output = finalizeOutput(rawContent);
@@ -926,7 +935,8 @@ export class FetchTool implements AgentTool<typeof fetchSchema, FetchToolDetails
 			throw new ToolAbortError();
 		}

-		const result = await renderUrl(url, effectiveTimeout, raw, signal);
+		const useKagiSummarizer = this.session.settings.get("fetch.useKagiSummarizer");
+		const result = await renderUrl(url, effectiveTimeout, raw, useKagiSummarizer, signal);
 		const truncation = truncateHead(result.content, {
 			maxBytes: DEFAULT_MAX_BYTES,
 			maxLines: FETCH_DEFAULT_MAX_LINES,
@@ -0,0 +1,96 @@
+import { afterEach, beforeEach, describe, expect, it, vi } from "bun:test";
+import * as fs from "node:fs";
+import * as os from "node:os";
+import * as path from "node:path";
+import { type SettingPath, Settings } from "@oh-my-pi/pi-coding-agent/config/settings";
+import type { ToolSession } from "@oh-my-pi/pi-coding-agent/tools";
+import { FetchTool } from "@oh-my-pi/pi-coding-agent/tools/fetch";
+import * as toolsManager from "@oh-my-pi/pi-coding-agent/utils/tools-manager";
+import * as kagi from "@oh-my-pi/pi-coding-agent/web/kagi";
+import type { LoadPageResult } from "@oh-my-pi/pi-coding-agent/web/scrapers/types";
+import * as scrapers from "@oh-my-pi/pi-coding-agent/web/scrapers/types";
+import { Snowflake } from "@oh-my-pi/pi-utils";
+
+describe("fetch tool Kagi summarization toggle", () => {
+	let testDir: string;
+
+	beforeEach(() => {
+		testDir = path.join(os.tmpdir(), `fetch-kagi-toggle-${Snowflake.next()}`);
+		fs.mkdirSync(testDir, { recursive: true });
+	});
+
+	afterEach(() => {
+		vi.restoreAllMocks();
+		fs.rmSync(testDir, { recursive: true, force: true });
+	});
+
+	const createSession = (overrides: Partial<Record<SettingPath, unknown>> = {}): ToolSession => {
+		const sessionFile = path.join(testDir, "session.jsonl");
+		return {
+			cwd: testDir,
+			hasUI: false,
+			getSessionFile: () => sessionFile,
+			getSessionSpawns: () => null,
+			settings: Settings.isolated({
+				"fetch.enabled": true,
+				...overrides,
+			}),
+		};
+	};
+
+	const mockLoadPage = () => {
+		const pageResponse = (url: string): LoadPageResult => {
+			if (url === "https://example.com") {
+				return {
+					ok: true,
+					status: 200,
+					contentType: "text/html",
+					finalUrl: "https://example.com",
+					content: "<html><body><h1>Example Domain</h1><p>Short sample content.</p></body></html>",
+				};
+			}
+
+			return {
+				ok: false,
+				status: 404,
+				contentType: "",
+				finalUrl: url,
+				content: "",
+			};
+		};
+
+		return vi
+			.spyOn(scrapers, "loadPage")
+			.mockImplementation(async (url: string, _options?: unknown) => pageResponse(url));
+	};
+
+	it("uses Kagi summarizer when enabled", async () => {
+		const session = createSession({ "fetch.useKagiSummarizer": true });
+		const tool = new FetchTool(session);
+		const loadPageSpy = mockLoadPage();
+		const summarizeSpy = vi.spyOn(kagi, "summarizeUrlWithKagi").mockResolvedValue("x".repeat(150));
+		vi.spyOn(toolsManager, "ensureTool").mockResolvedValue(undefined);
+		vi.spyOn(globalThis, "fetch").mockResolvedValue(new Response("blocked", { status: 500, statusText: "Blocked" }));
+
+		const result = await tool.execute("fetch-1", { url: "https://example.com" });
+
+		expect(loadPageSpy).toHaveBeenCalled();
+		expect(summarizeSpy).toHaveBeenCalledTimes(1);
+		expect(result.details?.method).toBe("kagi");
+	});
+
+	it("skips Kagi summarizer when disabled", async () => {
+		const session = createSession({ "fetch.useKagiSummarizer": false });
+		const tool = new FetchTool(session);
+		const loadPageSpy = mockLoadPage();
+		const summarizeSpy = vi.spyOn(kagi, "summarizeUrlWithKagi").mockResolvedValue("x".repeat(150));
+		vi.spyOn(toolsManager, "ensureTool").mockResolvedValue(undefined);
+		vi.spyOn(globalThis, "fetch").mockResolvedValue(new Response("blocked", { status: 500, statusText: "Blocked" }));
+
+		const result = await tool.execute("fetch-2", { url: "https://example.com" });
+
+		expect(loadPageSpy).toHaveBeenCalled();
+		expect(summarizeSpy).not.toHaveBeenCalled();
+		expect(result.details?.method).not.toBe("kagi");
+	});
+});