agentleFS
fitlab-aifitlab-ai/agent-infra/.agents/skills/entropy-check/SKILL.md

entropy-check

Skill88 starsChanged 20 days agoMITView on GitHub

周期性审查 agent-infra 仓库软熵增并输出结构化熵减审查报告。 当本仓库需要在发版前或维护窗口识别规则重叠、文档膨胀、死约定和版本散落风险时使用。

The file

Files
SKILL.md

熵减审查

行为边界 / 关键规则

  • 本技能是 agent-infra 仓库本地维护技能,不属于 update-agent-infra 分发模板。
  • 本技能不修改业务代码、规则、skill 或文档;只创建一份审查报告。
  • 报告写入 ~/.agent-infra/reports/agent-infra/entropy-check/entropy-check-YYYYMMDD-HHMMSS.md。
  • 发现项只给出证据与建议,不自动创建任务、不自动改文件。
  • 不确定的边界选择写入报告的 ## 人工裁决待办,不中途向用户提问。

执行步骤

1. 创建报告目录并做状态核对

创建输出目录:

mkdir -p "$HOME/.agent-infra/reports/agent-infra/entropy-check"

运行并记录以下命令,原文写入报告 ## 状态核对:

git status -s
git branch --show-current
date "+%Y-%m-%d %H:%M:%S%z" | sed 's/\([+-][0-9][0-9]\)\([0-9][0-9]\)$/\1:\2/'

2. 读取审查参考

执行审查前先读取:

  • reference/checklist.md
  • reference/report-template.md

3. 收集证据

按 checklist 收集证据,至少覆盖:

  • .agents/rules/issue-sync.md
  • .agents/rules/issue-pr-commands.md
  • .agents/rules/issue-fields.md
  • .agents/rules/pr-sync.md
  • .agents/rules/pr-checks-commands.md
  • .agents/rules/create-issue.md
  • .agents/skills/*/SKILL.md
  • templates/.agents/skills/**
  • 仓库源码、规则、Skill 与模板中的 TODO(compat)
  • package.json
  • .agents/.airc.json
  • 版本发布相关文档与脚本

使用 rg、find、wc -l、git status 等可复现命令收集证据,并把关键命令原文写入报告。

4. 执行语义审查

根据 reference/checklist.md 逐项判断是否存在软熵增:

  • 规则职责重叠或边界漂移
  • SKILL.md 膨胀且应拆分到 reference/
  • over-design、死约定或重复规则
  • 临时兼容 TODO 缺少删除对象、可验证条件,或删除条件已经满足
  • bilingual 命名约定混用且缺少边界说明
  • version 号散落导致漂移风险

每条发现必须包含证据、影响、建议和严重度。只有发版前必须处理的问题才标记为 release-blocking。

5. 写入报告

按 reference/report-template.md 创建报告:

report="$HOME/.agent-infra/reports/agent-infra/entropy-check/entropy-check-$(date "+%Y%m%d-%H%M%S").md"

报告必须保留结构化章节,便于发版流程引用报告路径并让维护者裁定后续任务。

6. 输出结论

输出报告路径和发现统计。若存在 release-blocking 发现,明确提示当前发版流程应先处理或由维护者裁定后继续。

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.