Token导航 LogoToken导航TokenDH.com
开发external-serviceclawhub未标认证来源可访问clear审计提醒

software-engineer软件工程师

Agent Skill

software-engineer 用于记录任务执行中的错误、用户纠正、经验和能力缺口,适合在 OpenClaw 中希望让 Agent 持续沉淀问题、修正和最佳实践时使用。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。

总安装

68,071

周安装

2,924

GitHub Stars

7

下载量

23,860
OpenClaw

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

MIT-0

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:software-engineer(软件工程师)
来源仓库:https://github.com/ivangdavila/software-engineer
安装命令:
openclaw skills install software-engineer
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 software-engineer

简介

使用干净的架构、正确的错误处理以及快速交付和正确构建之间的务实权衡来编写生产就绪的代码。

SKILL.md

name
Software Engineer
slug
software-engineer
version
1.0.0
homepage
https://clawic.com/skills/software-engineer
description
Write production-ready code with clean architecture, proper error handling, and pragmatic trade-offs between shipping fast and building right.
metadata
{"clawdbot":{"emoji":"👨‍💻","requires":{"bins":[]},"os":["linux","darwin","win32"]}}

When to Use

Agent needs to write, review, or refactor code. Handles implementation decisions, architecture trade-offs, and code quality across any language or framework.

Quick Reference

TopicFile
Code patternspatterns.md
Architecture decisionsarchitecture.md
Testing practicestesting.md

Core Rules

1. Read Before Write

  • Check existing code style, patterns, and conventions before writing new code
  • Respect the current stack — never swap libraries without explicit request
  • Match naming conventions, formatting, and project structure already in place

2. Code That Compiles

Every code block must:

  • Have correct imports for the actual library versions in use
  • Use APIs that exist in the project's dependency versions
  • Pass basic syntax checks — no placeholder // TODO: implement

3. Minimal First

  • Solve the specific problem, not hypothetical future problems
  • One abstraction when you have three concrete cases, not before
  • Features that might be needed → skip. Features that are needed → implement

4. Errors as First-Class Citizens

❌ catch (e) {}
❌ catch (e) { console.log(e) }
✅ catch (e) { logger.error('context', { error: e, input }); throw new DomainError(...) }
  • Typed errors over generic strings
  • Include context: what operation failed, with what input
  • Distinguish recoverable vs fatal errors

5. Boundaries and Separation

LayerContainsNever Contains
Handler/ControllerHTTP/CLI parsing, validationBusiness logic, SQL
Service/DomainBusiness rules, orchestrationInfrastructure details
Repository/AdapterData access, external APIsBusiness decisions

6. Explicit Trade-offs

When making architectural choices, state:

  • What you chose and why
  • What you traded away
  • When to revisit the decision

Example: "Using SQLite for simplicity. Trade-off: no concurrent writes. Revisit if >1 write/sec needed."

7. PR-Ready Code

Before delivering any code:

  • [ ] No dead code, commented blocks, or debug statements
  • [ ] Functions under 30 lines
  • [ ] No magic numbers — use named constants
  • [ ] Early returns over nested conditionals
  • [ ] Edge cases handled: null, empty, error states

Code Quality Signals

Senior code reads like prose:

  • Names explain "what" and "why", not "how"
  • A junior understands it in 30 seconds
  • No cleverness that requires comments to explain

The best code is boring:

  • Predictable patterns
  • Standard library over dependencies when reasonable
  • Explicit over implicit

Common Traps

TrapConsequencePrevention
Inventing APIsCode doesn't compileVerify method exists in docs first
Over-engineering3 hours instead of 30 minAsk: "Do I have 3 concrete cases?"
Ignoring contextSuggests wrong stackRead existing files before suggesting
Copy-paste without understandingHidden bugs surface laterExplain what the code does
Empty error handlingSilent failures in productionAlways log + type + rethrow
Premature abstractionComplexity without benefitYAGNI until proven otherwise

Pragmatic Shipping

Critical paths (do it right):

  • Authentication, authorization
  • Payment processing
  • Data integrity, migrations
  • Secrets management

Experimental paths (ship fast, iterate):

  • UI/UX features
  • Admin panels
  • Analytics
  • Anything unvalidated with users

Test for critical path: "Can this wake me at 3am or lose money?"

Security & Privacy

This skill does NOT:

  • Store any data externally
  • Make network requests
  • Access files outside the current project

All code suggestions are generated in context of the conversation.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

94.82%
按下载量换算22,624

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

未展示

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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