test(agent): prompt eval with fake model and real-model comparison harness

Co-Authored-By: Claude Opus 5.5 <[email protected]>
This commit is contained in:
rootandClaude Opus 5.5 committed 2026-09-30 18:47:15 +08:00
1 parent 21b7e8ff23
commit 425415b3ba
5 files changed
+705

No files matched your search

+4
View File
@@ -4,6 +4,7 @@ import type { FakeScript } from "@umt/agent";
import { makeTorrent } from "@umt/testkit";
import { createAcquisitionOrganizeModule } from "../../../src/modules/acquisition-organize/index.ts";
import { createAgentModule } from "../../../src/modules/agent/index.ts";
import type { AgentModuleOptions } from "../../../src/modules/agent/runner.ts";
import { authSettingsModule } from "../../../src/modules/auth-settings/index.ts";
import { createBrowserModule } from "../../../src/modules/browser/index.ts";
import type { BrowserModuleOptions } from "../../../src/modules/browser/service.ts";
@@ -32,6 +33,8 @@ export interface E2EModuleOptions {
webSearch?: WebSearchModuleOptions;
/** 浏览器模块注入(E14 视觉自动化:证据目录、动作后等待) */
browser?: BrowserModuleOptions;
/** 智能体模块的额外参数(E16:替换 / 捕获系统提示) */
agent?: Pick<AgentModuleOptions, "systemPrompt">;
}
export function buildModules(o: E2EModuleOptions): ServerModule[] {
@@ -71,6 +74,7 @@ export function buildModules(o: E2EModuleOptions): ServerModule[] {
sessionDir: o.sessionDir,
...(o.fakeScripts ? { fakeScripts: o.fakeScripts } : {}),
goalPollSec: 1,
...(o.agent ?? {}),
}),
// 浏览器能力(browser.*)随首批能力一起注册;浏览器服务由用例自己登记(E14 指向 mock camofox)
createBrowserModule(o.browser ?? {}),