codex-harness-engineering基于 OpenAI《Harness Engineering》实践的 Codex 工作规范。强调环境能力补齐、仓库即记录系统、严格分层架构、可观测可验证闭环、持续防漂移治理。
Install via ClawdBot CLI:
clawdbot install jimmypeng4ios/codex-harness-engineeringGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Generated May 24, 2026
An AI agent designs, codes, tests, and submits a PR for a multi-file feature in a web application. The agent self-reviews, responds to human feedback, and iterates until the PR is merged, ensuring high quality and adherence to architecture rules.
The agent scans dependencies, upgrades them across the codebase following the layered architecture, runs regression tests, and generates a PR with structured change logs and validation metrics.
The agent runs weekly background tasks to compare docs/ content with code structure, identifies outdated design docs, and submits pull requests to align documentation with current implementation using linter rules.
An agent queries LogQL and PromQL to detect performance degradation, analyzes root cause, and proposes code changes to improve latency or resource usage, all while maintaining strict architecture boundaries.
The agent refactors a legacy monolith into the prescribed Types->Config->Repo->Service->Runtime->UI layering, updates dependency injection and linter rules, and ensures no cross-boundary violations through automated tests.
Provide a managed service where AI agents handle feature development, code review, and maintenance for client software projects, following the harness engineering principles to ensure consistency and quality.
Sell or license the linter, CI checks, and documentation scaffold as a plugin for VS Code, JetBrains, or GitHub Actions, enabling teams to enforce the same processes in their own repositories.
Offer consulting engagements to help organizations adopt Codex Harness Engineering, including process design, toolchain setup, and team training, followed by ongoing support contracts.
💬 Integration Tip
Start by adopting the docs/ structure and the AGENTS.md map, then integrate the linter and CI rules step by step to avoid overwhelming the team.
Scored May 24, 2026
Control desktop applications on Windows — launch, close, focus, resize, move windows, simulate keyboard/mouse input, manage processes, control VSCode, read clipboard, and capture screen info. Use when the user wants to interact with any running program, switch windows, type text, press shortcuts, open files in VSCode, manage running processes, or get system display information.
Conduct rigorous, adversarial code reviews with zero tolerance for mediocrity. Use when users ask to "critically review" my code or a PR, "critique my code", "find issues in my code", or "what's wrong with this code". Identifies security holes, lazy patterns, edge case failures, and bad practices across Python, R, JavaScript/TypeScript, SQL, and front-end code. Scrutinizes error handling, type safety, performance, accessibility, and code quality. Provides structured feedback with severity tiers (Blocking, Required, Suggestions) and specific, actionable recommendations.
Coding style memory that adapts to your preferences, conventions, and patterns for consistent coding.
Pragmatic coding standards for writing clean, maintainable code — naming, functions, structure, anti-patterns, and pre-edit safety checks. Use when writing new code, refactoring existing code, reviewing code quality, or establishing coding standards.
Claude Code integration for OpenClaw. This skill provides interfaces to: - Query Claude Code documentation from https://code.claude.com/docs - Manage subagents and coding tasks - Execute AI-assisted coding workflows - Access best practices and common workflows Use this skill when users want to: - Get help with coding tasks - Query Claude Code documentation - Manage AI-assisted development workflows - Execute complex programming tasks
Plan, draft, version, and refine written content with enforced versioning and quality audits.