By rdcrystallab
Autonomous improvement engine for Claude Code. Runs an unbounded modify-verify-keep/discard loop against any mechanical metric. Includes planning wizard, security audit, shipping workflow, autonomous debugger, and autonomous fixer.
自主除錯迴圈——科學方法 + 自動研究迭代。找出所有 bug,而非只找一個。
自主修復迴圈——迭代修復錯誤直到歸零。每次迭代修一個,原子性操作,失敗自動還原。
互動式設定嚮導——從目標(Goal)產生範圍(Scope)、指標(Metric)、方向(Direction)與驗證指令(Verify)
自主安全審計——STRIDE 威脅模型 + OWASP Top 10 + 4 個對抗性角色的紅隊測試
通用發布工作流——透過結構化 8 階段工作流發布程式碼、內容、行銷、銷售、研究或任何事物
Claude Code 插件市集,提供各種擴展功能。
| 名稱 | 說明 | 類別 |
|---|---|---|
| damage-control | 損害控制安全系統,封鎖危險指令並保護敏感檔案 | security |
| drawio | 建立和編輯 draw.io 圖表,支援流程圖、架構圖、序列圖等 | visualization |
| excalidraw | 建立手繪風格 Excalidraw 圖表,支援 PNG 匯出 | visualization |
| github-kb | GitHub 本地知識庫管理員,維護本地 repo 索引,支援 clone、查詢與 gh CLI 搜尋 | productivity |
| playwright-cli | 瀏覽器自動化,支援網頁測試、表單填寫、截圖與資料擷取 | automation |
| tmux-fork | 開啟新的 tmux 視窗以並行執行 AI agents 或 CLI 指令 | automation |
| gh-tools | GitHub 工作流程指令,管理 issues、PR 與專案 | development |
| anvil | 以證據為優先的程式碼代理人,使用對抗性多模型審查與 SQL 追蹤驗證 | development |
| autoresearch | 自主改進引擎,執行無限循環的修改驗證流程,包含規劃精靈、安全審查與自動除錯 | productivity |
# 加入市集
claude /plugin marketplace add owner/agentic-plugins
# 安裝特定插件
claude /plugin install damage-control@agentic-plugins
# 加入整個市集
claude /plugin add /path/to/agentic-plugins
# 或安裝單一插件
claude /plugin add /path/to/agentic-plugins/plugins/damage-control
agentic-plugins/
├── .claude-plugin/
│ └── marketplace.json # 市集註冊
├── plugins/
│ ├── damage-control/ # 安全插件
│ ├── drawio/ # draw.io 圖表
│ ├── excalidraw/ # Excalidraw 圖表
│ ├── github-kb/ # GitHub 知識庫
│ ├── playwright-cli/ # 瀏覽器自動化
│ ├── tmux-fork/ # tmux 並行執行
│ ├── gh-tools/ # GitHub 工具
│ ├── anvil/ # 程式碼驗證代理人
│ └── autoresearch/ # 自主改進引擎
└── README.md
每個插件目錄包含:
plugin-name/
├── .claude-plugin/
│ └── plugin.json # 插件元數據
├── agents/ # Agent 定義(選用)
├── skills/ # Skill 定義(選用)
├── commands/ # Slash 指令(選用)
├── hooks/ # 事件鉤子(選用)
└── README.md
新增插件時請遵循:
plugins/ 下建立插件目錄.claude-plugin/plugin.json 元數據README.md.claude-plugin/marketplace.json 中註冊MIT
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 claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub rdcrystallab/agentic-plugins --plugin autoresearch使用 draw.io 建立和編輯 XML 格式的圖表。透過智慧邊緣路由與佈局最佳化,產生流程圖、架構圖、序列圖等。
損害控制 (Damage Control) 安全系統 - 透過 PreToolUse 鉤子封鎖危險指令並保護敏感檔案
Browser automation with playwright-cli for web testing, form filling, screenshots, and data extraction
Evidence-first coding agent. Verifies before presenting. Attacks its own output with adversarial multi-model review, IDE diagnostics, and SQL-tracked verification to ensure code quality.
使用 Excalidraw JSON 格式建立手繪風格圖表。產生架構圖、流程圖等視覺化圖表,並支援 PNG 匯出。
Autonomous improvement engine for Claude Code. Runs an unbounded modify-verify-keep/discard loop against any mechanical metric. 10 subcommands: plan, debug, fix, security, ship, scenario, predict, learn, and reason.
Automated development loop and ad-hoc workflow generation for orchestrating AI coding agents
Meta-cognitive tools for Claude Code self-improvement. Learn from feedback, optimize configuration, and evolve your AI development workflow.
Autonomous experiment loop that optimizes any file by a measurable metric. 5 slash commands, 8 evaluators, configurable loop intervals (10min to monthly).
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
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.