By HerbertGao
Orchestrated OpenSpec apply — the main agent groups pending tasks by scope, dispatches a subagent per group (parallel then serial by dependency), and only manages OpenSpec state and reviews; it never writes implementation code itself. Ships as /opsx:apply-subagent plus its SOT skill.
HerbertGao 的自托管 Claude Code skills marketplace。集中管理自建 skill,避免散落丢失,并支持任意机器一键安装/更新。
/plugin marketplace add HerbertGao/claude-skills
/plugin install review-loop@claude-skills
/plugin install opsx@claude-skills
更新:/plugin marketplace update claude-skills 后重装对应 plugin。
注:
opsx需要openspec-cnCLI(plugin 不附带,请自行安装)。若你此前已在~/.claude手动放过同名 command/skill,安装本 plugin 会产生重复,二选一即可。
| Plugin | 说明 |
|---|---|
review-loop | 对本次提案与代码变更跑对抗性 review 自动循环:Codex + Code Reviewer + Reality Checker 三方并行 → 合并去重 triage → 派 Minimal Change Engineer 修复 → 重新 review,直到放行(APPROVE/CLEAR;Codex 结构性未跑完时 APPROVE-DEGRADED)或到可配置轮数上限。常配合内置 /goal 长跑。 |
opsx | 以编排模式实现 OpenSpec 变更:主 agent 按范围(子项目 / 模块 / 技术栈)分组待处理任务,先并行后串行派发 subagent 开发,自己只跑 openspec-cn 状态管理与 review,不亲自写实现代码。提供 /opsx:apply-subagent 命令 + 其 SOT skill。需要 openspec-cn CLI。 |
claude-skills/
├─ .claude-plugin/
│ └─ marketplace.json # marketplace 清单
├─ review-loop/ # 一个 plugin(纯 skill)
│ ├─ .claude-plugin/
│ │ └─ plugin.json # plugin 清单
│ └─ skills/
│ └─ review-loop/
│ └─ SKILL.md # skill 本体(SOT)
└─ opsx/ # 一个 plugin(命令 + skill)
├─ .claude-plugin/
│ └─ plugin.json # plugin 清单
├─ commands/
│ └─ apply-subagent.md # /opsx:apply-subagent 薄命令
└─ skills/
└─ openspec-apply-change-subagent/
└─ SKILL.md # skill 本体(SOT)
新增 skill:在仓根加一个同构的 <plugin>/ 目录,并在 .claude-plugin/marketplace.json 的 plugins[] 追加一条。
MIT
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub herbertgao/claude-skills --plugin opsxPersistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Convert QQ Music .ogg/.oggh files on macOS to .mp3 with embedded cover art (from Resource Fork ICNS) and metadata. Provides /ogg2mp3:* slash commands and a Skill description for natural-language invocation. All operations run unprivileged.
Manage multiple WeChat instances on macOS via natural language. Provides /double-wechat:* slash commands and a Skill description; all writes run unprivileged.
Operate the user's own X (Twitter) Likes via natural language — list, download media, check auth health. Backed by a local x_likes_downloader MCP server.
Adversarial review loop that iterates your proposal and diff to a clean pass — parallel Codex + Code Reviewer + Reality Checker review, merge/triage, minimal-change fixes, re-review until APPROVE/CLEAR or a configurable round cap.