claude-code-codex-strongest
AGENTS.md1 starsChanged 6 months agoLicence not identifiedView on GitHub
本文件是本项目及其衍生配置的共享工作协议。目标不是让 AI 取代人类，而是让人类和 AI 各自承担最擅长、最可靠的部分：AI 扩大探索和执行能力，人类把握目标、意义、取舍、深层理解、关键感知、纠错和最终责任。 规则必须服务于工作，不得变成官僚流程。计划、汇报和确认的详细程度要与任务风险和复杂度相称。 能力边界会随模型、工具和任务变化。下面是默认分工，不是“某一方永远完胜”的绝对断言；高影响任务必须用证据重新校准。 正式实现前，必须完成以下闭环： 理解 → 调研 → 写计划 → 人类确认或明确授权 → 分批执行 → 持续同步 → 验证 → 通俗交付 不要为了形式给一行改动创建长文档，也不要用“任务很大”作为跳过理解和计划的理由。 默认将正式计划写入仓库根目录的 PLAN.md；仓库已有计划目录或同一时间有多个独立任务时，使用 plans/<任务名>.md。不要为同一个任务维护互相冲突的多个计划。 计划必须具体到可以执行和检查，不能只有“实现功能”“优化代码”之类的空话。交互模式下，计划写完后先停下来让人类确认；无人值守模式下，计划是共享状态和恢复依据，不是可以悄悄绕过的手续。 如果新事实证明计划错误、范围需要扩大或方向需要改变：停止继续扩张，说明新旧差异，更新计划并重新取得授权。不要事后把已经做过的工作包装成原计划的一部分。
- Reads credentials
In this repository36 other files it gives its agents.
- .codex/AGENTS.mdAGENTS.md
- commands/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- brainstormingSkill
- condition-based-waitingSkill
- defense-in-depthSkill
- dispatching-parallel-agentsSkill
- document-handlingSkill
- executing-plansSkill
- extracting-from-imagesSkill
- finishing-a-development-branchSkill
- incremental-context-buildingSkill
- lean-contextSkill
- mcp-builderSkill
- prompt-engineeringSkill
- receiving-code-reviewSkill
- requesting-code-reviewSkill
- root-cause-tracingSkill
- skill-creatorSkill
- subagent-driven-developmentSkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- using-claude-apiSkill
- using-context7Skill
- using-cronSkill
- using-deepwikiSkill
- using-git-worktreesSkill
- using-mcp-memorySkill
- using-memorySkill
- using-playwrightSkill
- using-sequential-thinkingSkill
- using-superpowersSkill
- using-task-listSkill
- verification-before-completionSkill
- writing-plansSkill
- writing-skillsSkill
The file
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.