Token导航 LogoToken导航TokenDH.com
研究检索只读clawhub未标认证来源可访问clear审计通过

task-retrospective任务回顾

Agent Skill

task-retrospective 用于查找、检索和筛选相关信息,适合在 OpenClaw 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

1,536

周安装

64

GitHub Stars

公开资料未说明

下载量

512
OpenClaw

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

2

许可证

MIT-0

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:task-retrospective(任务回顾)
来源仓库:https://github.com/charlie-morrison/task-retrospective
安装命令:
openclaw skills install task-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。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install task-retrospective

简介

在任务结束后进行结构化的自我评估,以分析效率、准确性、方法质量并提取模式以提高未来的绩效。

SKILL.md

Task Retrospective

Structured self-evaluation for AI agents after completing tasks. Analyze what worked, what failed, and extract reusable patterns to improve future performance. Use after completing complex tasks, debugging sessions, or multi-step workflows.

Usage

Run a retrospective on the task I just completed.

Or with specific context:

Retrospective: [task description]. 
Outcome: [success/partial/failure].
Time spent: [duration].
What surprised me: [unexpected findings].

How It Works

  1. Reconstruct — review the task timeline (steps taken, tools used, decisions made)
  2. Evaluate — score each phase on efficiency, accuracy, and approach quality
  3. Extract — identify reusable patterns, anti-patterns, and decision heuristics
  4. Record — generate a structured retrospective for future reference

Evaluation Dimensions

Efficiency

  • Were there unnecessary steps or dead ends?
  • Could tool calls have been batched or parallelized?
  • Was the research phase too long or too short?

Accuracy

  • Was the final output correct and complete?
  • Were there false starts or incorrect assumptions?
  • Did the solution match the actual requirements?

Approach Quality

  • Was the problem decomposition effective?
  • Were the right tools chosen for each step?
  • Would a different strategy have been faster?

Learning Extracted

  • What new patterns can be reused?
  • What anti-patterns should be avoided?
  • What domain knowledge was gained?

Output Format

## Task Retrospective

### Summary
[1-2 sentences: what was the task, what was the outcome]

### Timeline
| Phase | Duration | Verdict |
|-------|----------|---------|
| Research | Xm | Efficient / Too long / Insufficient |
| Planning | Xm | Good / Skipped / Over-planned |
| Execution | Xm | Clean / Had rework / Multiple attempts |
| Validation | Xm | Thorough / Skipped / Caught issues |

### What Worked
- [Pattern that should be repeated]

### What Didn't Work
- [Anti-pattern to avoid] → [Better alternative]

### Reusable Patterns
- **Pattern name**: [Description of when and how to apply]

### Key Decisions
- [Decision point] → [Choice made] → [Outcome: good/bad/neutral]

### Improvement Actions
- [ ] [Specific action to improve future performance]

Advanced Usage

Compare Approaches

Compare my approach to [task] with the ideal approach. 
What I did: [steps].
What I should have done: [if known].

Pattern Library

Over time, retrospectives build a pattern library:

Review my last 5 retrospectives. What recurring patterns emerge?
Which improvement actions have I actually followed through on?

Team Retrospective

Run a retrospective on this multi-agent workflow.
Agents involved: [list].
Handoff points: [where work transferred between agents].
Bottlenecks: [where things slowed down].

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

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

OpenClaw

88.67%
按下载量换算454

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills