npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@ai-content-space/loopx

v0.10.1

Published

Docs-first engineering discipline for agentic coding assistants

Readme

English

loopx 把工程纪律编译成文档,并用证据证明它们有效。核心交付物是一份安装进 host guidance(AGENTS.md / CLAUDE.md)的简短 working agreement,加上三个 产出文档的 skills。执行属于模型和宿主运行时:loopx 不提供执行 runtime、评审 流水线或每轮 hook;可选的 $exec 只是消费 ready plan2exec 文档的宿主原生 subagent playbook。

日常工作在 working agreement 之下保持 prompt-first:先检查需求和相关代码, 主动找出缺失的实质决策,明确后再开始依赖这些决定的实现。清楚且边界明确的 请求不需要额外确认。保持最小正确改动,检查边界条件并完成新鲜验证, 没有明确请求绝不做 Git 处置。

三个 canonical workflow intents 都产出文档:

  • clarify 在实现前找出需求缺口,集中询问独立决定,依赖问题等前提明确后追问, 产出带可测试 AC-* / TC-* 锚点的需求契约。
  • spec 把长期有效的产品、兼容、数据、安全或架构决策固化为带 D-* 锚点的 设计文档。
  • plan2exec 只在明确要求计划、审批边界、中断恢复或持久协调时写一份 lean plan 文档,由执行 agent 自己按文档执行。该名称与 agent 内建 Plan 模式明确 区分。
  • exec 可选地通过 leaf subagent 执行一份 ready plan,由顶层模型负责审查、 顺序集成、冲突处理与验证。

debug 统一负责排查和修复,替代 issue、fix。只查原因时交付诊断, 要求修复时继续修改和验证,默认不建台账。已有台账仍可读取,并保留原有范围 和恢复检查。tdd、debug、 api-designer、generate-api-docs、architecture-designer、 sql-style、cli-developer、lancet、prompt-lint 等支持 skills 仍是 lenses,不是 workflow 状态。prompt-lint 只评估任务定义质量,不执行被评估的 提示词。

为什么 docs-first

支撑这个设计的四臂 benchmark(evals/benchmark/RESULTS.md)表明:对前沿模型, 仅 working agreement 一份文档就在停下纪律上与完整的 v0.7 治理运行时打平 (相对裸模型 +65pp),token 成本只有三分之一;而运行时没有带来能力增益。 所以 loopx 交付文档本身,以及让文档保持诚实的证据流水线:evals/drills/ 对每条纪律条款做压力测试,evals/benchmark/ 对照裸模型与 docs-only 对照组 度量通过率与 token 经济。

安装

npm install -g @ai-content-space/loopx
loopx install-skills --target all --yes
loopx doctor

先预览将要安装的文件:

loopx install-skills --target all --dry-run

完整 CLI 与安装细节见 CLI Reference。

在 Agent 中使用

$clarify <ambiguous-request>
$spec <decision-heavy-change>
$plan2exec <approved-source-or-planning-request>
$exec <ready-plan> [model=<id>] [reasoning_effort=<level>] [max_workers=<n>]

其余一切都是 working agreement 之下的普通模型工作。每次完成声明都需要新鲜 验证;高风险 diff 的独立评审是 working agreement 的条款,由宿主原生 subagent 执行。

上下文规则

docs/loopx/specs/ 保存长期有效、具有约束力的 repo context。 docs/loopx/decisions/docs-first-pivot.md 记录当前架构决策。 docs/archive/ 只保存历史,不属于当前权威,并从默认检索中排除。 .loopx/memory/MEMORY.md 是建议性的 curated memory。当前用户指令和已批准的 source documents 优先级更高。

维护者

发布前运行确定性治理门:

node scripts/verify-skills.mjs

Package 与 plugin manifest 版本跟随 npm 发布。Skill metadata.version 独立, 只在内容或行为契约变化的 skill 上递增。