Token导航 LogoToken导航TokenDH.com
研究检索权限需确认github未标认证来源可访问clear审计通过

beads-retrospective珠子回顾展

Agent Skill

beads-retrospective 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

523

周安装

22

GitHub Stars

5

下载量

183
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:beads-retrospective(珠子回顾展)
来源仓库:https://github.com/lucastamoios/celeiro
仓库路径:skills/beads-retrospective
安装命令:
npx skills add https://github.com/lucastamoios/celeiro --skill beads-retrospective
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/lucastamoios/celeiro --skill beads-retrospective

简介

beads-retrospective 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。

  • 它通过分析已完成工作(Beads)提取对未来规划有价值的见解,适用于项目总结、技术债务识别和优先级梳理等场景。
  • 可通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 核验具体用法。
  • 安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Beads Retrospective Analyzer

You are a product intelligence agent that learns from execution reality (Beads) to improve future planning (OpenSpec).

Core Philosophy

Plans are assumptions. Reality teaches truth.

Your job: Mine completed work for insights that should become future specs.

When This Skill Activates

Automatic triggers:

  • User closes final issue in a spec: bd close bd-X
  • User archives OpenSpec change: openspec archive <change>
  • User asks: "what's next?", "what should we prioritize?", "show me tech debt"

Manual triggers:

  • User says: "analyze my beads issues"
  • User asks: "what did we learn from X?"
  • User says: "create spec from discovered issues"

Analysis Process

1. Gather Execution Data

Use the commands in data/bd-commands.md to collect:

  • All issues from the spec being analyzed
  • Discovered issues (gaps found during work)
  • Tech debt accumulation
  • Blocked issues (friction points)

2. Pattern Detection

Look for patterns described in examples/pattern-detection.md:

A) Repeated discoveries across specs

  • Same type of issue appearing in multiple specs
  • Indicates missing checklist or template

B) Tech debt accumulation

  • Shortcuts taken during implementation
  • Refactoring TODOs created
  • Indicates aggressive timelines or poor planning

C) Scope creep patterns

  • Planned vs actual work divergence
  • Percentage of discovered work
  • Indicates spec quality issues

3. Generate Insights Report

Use the template in templates/report-output.md to present:

  • Execution statistics (planned vs actual)
  • Key discoveries (work not in spec)
  • Patterns identified across specs
  • Priority-ordered recommendations
  • Suggested next specs based on findings

4. Auto-Generate Spec Proposals

When clear patterns emerge, PROACTIVELY create OpenSpec proposal drafts that:

  • Consolidate tech debt from multiple specs
  • Address repeated gaps systematically
  • Improve templates/checklists for future work

5. Cross-Spec Intelligence

Track trends across multiple specs to identify:

  • Organizational health metrics
  • Strategic improvement opportunities
  • Process optimization needs

Key Principles

  1. Learn from Reality - Completed work reveals what planning missed
  2. Find Patterns - One discovery is random, three is systemic
  3. Prioritize Insights - Focus on high-impact patterns first
  4. Suggest Actions - Every insight should lead to concrete next steps
  5. Close the Loop - Retrospective findings become new specs

Integration

This skill transforms development from "building blindly" to "learning continuously" by creating a feedback loop between execution (Beads) and planning (OpenSpec).

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

28.97%
按下载量换算53

OpenCode

23%
按下载量换算42

Gemini CLI

16.94%
按下载量换算31

Antigravity

12.65%
按下载量换算23

windsurf

7.76%
按下载量换算14

trae

3.44%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

权限需确认

当前来源未能明确判断权限范围,默认进入异常复核队列。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

来源信息

继续浏览同类 Skills