Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计提醒

testing-flow测试流程

Agent Skill

用于辅助测试设计、自动化测试、用例整理和回归验证。它适合让 Agent 编写单元测试、端到端测试、测试计划或根据失败日志定位问题。使用时需要确认项目测试框架、运行命令和夹具数据,避免为了通过测试而改坏真实逻辑;涉及浏览器或外部服务时,应区分本地模拟、测试环境和生产环境。

总安装

324

周安装

13

GitHub Stars

1

下载量

105
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:testing-flow(测试流程)
来源仓库:https://github.com/robomotionio/agent-skills
仓库路径:skills/testing-flow
安装命令:
npx skills add https://github.com/robomotionio/agent-skills --skill testing-flow
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/robomotionio/agent-skills --skill testing-flow

简介

用于端到端业务流程的测试场景建模。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

  • 适合绘制用户旅程图并生成关键路径测试用例。
  • 支持多系统集成点的契约验证与异常流覆盖。
  • 应优先覆盖高频核心功能再扩展边缘情况。
  • testing-flow 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Testing a Robomotion Flow

Run behavioral tests for a Robomotion flow. Uses @robomotion/sdk/testing + bun test.

For pspec schema validation, use validating-flow.

Purpose

  • Execute *.test.ts files with bun test
  • Author new tests with FlowTester, MockRegistry, SubflowTester
  • Inspect structure, mock services, trace execution, exercise Function-node logic

Workflow

  1. Check for testsls path/to/flow/*.test.ts
  2. Runcd path/to/flow && bun test
  3. Fix failures — diagnose, edit, re-run

Quick Start

Create main.test.ts next to main.ts:

import { describe, test, expect, beforeAll, afterAll } from 'bun:test';
import { FlowTester, MockRegistry } from '@robomotion/sdk/testing';

describe('My Flow', () => {
  let tester: FlowTester;

  beforeAll(async () => {
    tester = await FlowTester.load('./main.ts');
  });

  afterAll(() => {
    MockRegistry.clear();
  });

  test('validates', () => {
    expect(tester.validate().valid).toBe(true);
  });

  test('has expected node count', () => {
    expect(tester.getNodes().length).toBe(10);
  });
});

API Map

Load only the reference file(s) you need. Each is self-contained and lives inside this skill under ./reference/.

AreaWhat's in itReference
Load/inspect flows, run checkpoints, analyze paths and complexityFlowTester, analyzeFlow, findPaths, generateDotGraph./reference/flow-tester.md
Mock external services (per-node, per-type, helpers, call tracking)MockRegistry, mockResponse/Error/Sequence/Conditional./reference/mock-registry.md
Test Function-node JavaScript in isolationextractFunc, runFunc, testFunctionWithCases, analyzeFunctionNode./reference/function-extraction.md
Run subflows in isolation; discover them in a flowSubflowTester, discoverSubflows, loadAllSubflows./reference/subflow-tester.md
Unit vs real-service mode via env varsisIntegrationMode, hasCredentials, createTestHelper./reference/integration-mode.md

Complete Example

import { describe, test, expect, beforeAll, afterAll, afterEach } from 'bun:test';
import {
  FlowTester, MockRegistry, extractFunc,
  isIntegrationMode, hasCredentials, analyzeFlow,
} from '@robomotion/sdk/testing';

describe('Reddit to WordPress Flow', () => {
  let tester: FlowTester;

  beforeAll(async () => { tester = await FlowTester.load('./main.ts'); });
  afterEach(() => MockRegistry.resetCalls());
  afterAll(() => MockRegistry.clear());

  describe('Structure', () => {
    test('validates', () => expect(tester.validate().valid).toBe(true));
    test('all branches terminate', () => expect(tester.allBranchesTerminate()).toBe(true));
    test('loops have Goto->Label', () => {
      expect(tester.findLoops().every(l => l.hasGotoLabel)).toBe(true);
    });
    test('no dead ends', () => {
      expect(analyzeFlow(tester).deadEnds).toHaveLength(0);
    });
  });

  describe('Transform Logic', () => {
    test('formats Reddit post', () => {
      const transform = extractFunc(tester, '2bf967');
      const out = transform({ post: { title: 'Test', selftext: 'Body', subreddit: 'programming' } });
      expect(out.wpTitle).toContain('Test');
    });
  });

  describe('With Mocked Services', () => {
    beforeAll(() => {
      MockRegistry.registerNode('c8f412', async () => ({
        posts: [{ title: 'Mock 1', selftext: 'c1', score: 100 }],
      }));
      MockRegistry.register('Core.Net.HttpRequest', async (p, m, nodeId) => {
        console.warn(`Unmocked HTTP in ${nodeId}`);
        return { statusCode: 200, body: {} };
      });
    });

    test('fetches posts', async () => {
      const res = await tester.runUntil('c5d901', {
        input: { subreddit: 'programming', limit: 10 },
      });
      expect(res.msg.posts).toHaveLength(1);
      expect(MockRegistry.getCallCount('c8f412')).toBe(1);
    });
  });

  describe('Integration', () => {
    test('real WordPress posting', async () => {
      if (!isIntegrationMode() || !hasCredentials('wordpress')) return;
      const res = await tester.runWithTrace({
        input: { subreddit: 'programming', postLimit: 1 },
        useRealServices: true,
      });
      expect(res.output.publishedUrl).toBeDefined();
    });
  });
});

Tips

  • Use node-ID mocks for precise control; type mocks as catch-all.
  • Verify mock inputs with getLastCall().
  • Clear mocks in afterAll() to avoid pollution across files.
  • INTEGRATION=1 bun test switches to real services.

Related Skills

  • validating-flow — schema validation (property names, ports, node types)
  • creating-flow — generate TypeScript
  • running-flow — execute on a robot

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

41%
按下载量换算43

Claude

29.46%
按下载量换算31

Cursor

17.83%
按下载量换算19

Gemini CLI

9.36%
按下载量换算10

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills