Token导航 LogoToken导航

Agent Skill · 研究检索

workflow-patterns(工作流程模式)

workflow-patterns 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

workflow-patterns 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为只读;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 4,082。

总安装 36,317周安装 1,560GitHub Stars 公开资料未说明下载 12,730核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

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:workflow-patterns(工作流程模式)
来源仓库:https://github.com/wpank/workflow-patterns
安装命令:
openclaw skills install workflow-patterns
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install workflow-patterns

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

使用 TDD、阶段检查点和结构化提交来系统化任务实施。在继续之前通过红绿重构周期、80% 覆盖率门和验证协议确保质量。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install workflow-patterns;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

workflow-patterns 怎么安装?

在终端执行 openclaw skills install workflow-patterns。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

workflow-patterns 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

workflow-patterns 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

workflow-patterns 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

workflow-patterns 用的人多吗?

公开数据是累计安装 36,317 次,近一周安装 1,560 次,下载 12,730 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

workflow-patterns 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
workflow-patterns

工作流程模式 · 本页

36,317—只读MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
workflow-patterns
model
standard
version
2.0.0
description
>
tags
[tdd, workflow, quality, testing, git, checkpoints, implementation]

Workflow Patterns

Implement tasks systematically using TDD (Test-Driven Development) with phase checkpoints and verification protocols. Ensures quality at every step.

Installation

OpenClaw / Moltbot / Clawbot

npx clawhub@latest install workflow-patterns

WHAT This Skill Does

Provides a structured approach to implementing tasks:

  • TDD cycle (red → green → refactor) for each task
  • Quality gates (tests, coverage, linting) before marking complete
  • Phase checkpoints requiring user approval
  • Git commits with rich metadata for traceability

WHEN to Use

Use for:

  • Implementing features from a plan
  • Following TDD methodology
  • Tasks requiring quality verification
  • Projects with coverage requirements
  • Team workflows needing traceability

Skip for:

  • Quick fixes or trivial changes
  • Exploratory prototyping
  • Projects without test infrastructure

Keywords: TDD, implementation, testing, coverage, checkpoints, verification, red-green-refactor

The TDD Task Lifecycle

11 steps for each task:

Step 1: Select Next Task

Read the plan and identify the next pending [ ] task. Select tasks in order within the current phase. Do not skip ahead.

Step 2: Mark as In Progress

Update the plan to mark the task as [~]:

- [~] **Task 2.1**: Implement user validation

Step 3: RED — Write Failing Tests

Write tests that define expected behavior before implementation:

  • Create test file if needed
  • Cover happy path
  • Cover edge cases
  • Cover error conditions
  • Run tests — they should FAIL
def test_validate_email_valid():
    user = User(email="test@example.com")
    assert user.validate_email() is True

def test_validate_email_invalid():
    user = User(email="invalid")
    assert user.validate_email() is False

Step 4: GREEN — Implement Minimum Code

Write the minimum code to make tests pass:

  • Focus on making tests green, not perfection
  • Avoid premature optimization
  • Keep implementation simple
  • Run tests — they should PASS

Step 5: REFACTOR — Improve Clarity

With green tests, improve the code:

  • Extract common patterns
  • Improve naming
  • Remove duplication
  • Simplify logic
  • Run tests after each change — must stay GREEN

Step 6: Verify Coverage

Check test coverage meets the 80% target:

pytest --cov=module --cov-report=term-missing

If coverage is below 80%:

  • Identify uncovered lines
  • Add tests for missing paths
  • Re-run coverage check

Step 7: Document Deviations

If implementation deviated from plan or added dependencies:

  • Update tech-stack.md with new dependencies
  • Note deviations in plan task comments
  • Update spec if requirements changed

Step 8: Commit Implementation

Create focused commit:

git commit -m "feat(user): implement email validation

- Add validate_email method to User class
- Handle empty and malformed emails
- Add comprehensive test coverage

Task: 2.1"

Step 9: Update Plan with SHA

Mark task complete with commit SHA:

- [x] **Task 2.1**: Implement user validation `abc1234`

Step 10: Commit Plan Update

git commit -m "docs: update plan - task 2.1 complete"

Step 11: Repeat

Continue to next task until phase is complete.

Phase Completion Protocol

When all tasks in a phase are complete:

1. Identify Changed Files

git diff --name-only <last-checkpoint-sha>..HEAD

2. Ensure Test Coverage

For each modified file:

  • Verify tests exist for new/changed code
  • Run coverage for modified modules
  • Add tests if coverage < 80%

3. Run Full Test Suite

pytest -v --tb=short

All tests must pass.

4. Generate Verification Checklist

## Phase 1 Verification

- [ ] User can register with valid email
- [ ] Invalid email shows appropriate error
- [ ] Database stores user correctly

5. WAIT for User Approval

Present checklist:

Phase 1 complete. Please verify:
1. [x] Test suite passes (automated)
2. [x] Coverage meets target (automated)
3. [ ] Manual verification items (requires human)

Respond with 'approved' to continue.

Do NOT proceed without explicit approval.

6. Create Checkpoint Commit

git commit -m "checkpoint: phase 1 complete

Verified:
- All tests passing
- Coverage: 87%
- Manual verification approved"

7. Record Checkpoint SHA

Update plan checkpoints table:

## Checkpoints

| Phase   | SHA     | Date       | Status   |
|---------|---------|------------|----------|
| Phase 1 | def5678 | 2025-01-15 | verified |
| Phase 2 |         |            | pending  |

Quality Gates

Before marking any task complete:

GateRequirement
TestsAll existing tests pass, new tests pass
CoverageNew code has 80%+ coverage
LintingNo linter errors
TypesType checker passes (if applicable)
SecurityNo secrets in code, input validation present

Git Commit Format

<type>(<scope>): <subject>

<body>

Task: <task-id>

Types:

  • feat — New feature
  • fix — Bug fix
  • refactor — Code change without feature/fix
  • test — Adding tests
  • docs — Documentation
  • chore — Maintenance

Handling Deviations

Scope Addition

Discovered requirement not in spec:

  • Document in spec as new requirement
  • Add tasks to plan
  • Note addition in task comments

Scope Reduction

Feature deemed unnecessary:

  • Mark tasks as [-] (skipped) with reason
  • Update spec scope section
  • Document decision rationale

Technical Deviation

Different approach than planned:

  • Note deviation in task comment
  • Update tech-stack.md if dependencies changed
  • Document why original approach was unsuitable
- [x] **Task 2.1**: Implement validation `abc1234`
  - DEVIATION: Used library instead of custom code
  - Reason: Better edge case handling
  - Impact: Added email-validator to dependencies

Error Recovery

Tests Fail After GREEN

  1. Do NOT proceed to REFACTOR
  2. Identify which test started failing
  3. Revert to last known GREEN state
  4. Re-approach the implementation

Checkpoint Rejected

  1. Note rejection reason in plan
  2. Create tasks to address issues
  3. Complete remediation tasks
  4. Request checkpoint approval again

Blocked by Dependency

  1. Mark task as [!] with blocker description
  2. Check if other tasks can proceed
  3. Document expected resolution

Task Status Symbols

SymbolMeaning
[ ]Pending
[~]In progress
[x]Complete
[-]Skipped
[!]Blocked

Best Practices

  1. Never skip RED — Always write failing tests first
  2. Small commits — One logical change per commit
  3. Immediate updates — Update plan right after task completion
  4. Wait for approval — Never skip checkpoint verification
  5. Coverage discipline — Don't accept below target
  6. Sequential phases — Complete phases in order
  7. Document deviations — Note any changes from plan
  8. Clean state — Each commit leaves code working

NEVER Do

  1. NEVER skip the RED phase — writing tests first is non-negotiable in TDD
  2. NEVER proceed past checkpoints without approval — wait for explicit user confirmation
  3. NEVER commit code that doesn't pass tests — every commit must be a working state
  4. NEVER accept coverage below 80% — add tests until threshold is met
  5. NEVER hide deviations from the plan — document all changes from original spec
  6. NEVER skip phases or reorder them — phases are sequential for a reason
  7. NEVER forget to record commit SHAs — traceability requires linking tasks to commits

平台分布

OpenClaw

97.03%
按下载量换算12,352

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills