技能 w-retro

📦

w-retro

Name: w-retro
Author: blockmatic

内容修订 r1 安全

Review Completed Work and Improve Next Time

Teams often finish tasks without recording what changed, what evidence was missing, or what should improve. This skill produces a concise, task-specific retrospective with up to three practical next steps.

支持: Claude Codex Code(CC)

🥉 78 青铜

使用我的智能体安装

将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。

智能体请求

Review the Skillstore skill "w-retro" from https://skillstore.io/skills/blockmatic-w-retro.md and its manifest at https://skillstore.io/api/skills/blockmatic-w-retro/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智能体仍应展示其计划，并按安全策略请求必要确认。

Agent 可读资源

当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。

Markdown 详情 GET /skills/blockmatic-w-retro.md 已签名 manifest GET /api/skills/blockmatic-w-retro/manifest 已签名 lockfile GET /api/skills/blockmatic-w-retro/lockfile

测试它

正在使用“w-retro”。 We needed to add export filtering. The filter shipped and unit tests passed, but no browser test was run.

预期结果:

Requested versus shipped: Export filtering was added and unit-tested. Evidence gap: Browser behavior was not checked. Next change: Add one browser test for filter selection and downloaded results.

正在使用“w-retro”。 The task was to update the onboarding guide. The guide changed, but the examples were not checked against the current interface.

预期结果:

Requested versus shipped: The onboarding guide was updated, but example accuracy was not verified.
Process gap: The task lacked a user-interface consistency check.
Next changes: Review examples against the current interface and add a lightweight documentation checklist.

安全审计

安全

v1 • 2026/9/20 打开版本化报告

The scanned skill is a 12-line, chat-only retrospective checklist with no executable code or detected security patterns. No prompt-injection, data-exfiltration, or business-logic abuse evidence was found in SKILL.md.

已扫描文件

分析行数

复核项

已忽略误报

最新一次完成的静态与语义审计未检测到已确认的安全发现。这并不能证明该 Skill 没有副作用。

审计者: codex

分享与引用此报告

分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。

打开版本化报告

复制报告链接

https://skillstore.io/skills/blockmatic-w-retro/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/blockmatic-w-retro/security.svg)](https://skillstore.io/skills/blockmatic-w-retro?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/blockmatic-w-retro?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/blockmatic-w-retro/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/blockmatic-w-retro.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

学术引用 (APA · BibTeX · CFF)

APA 引用

blockmatic. (2026). w-retro security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/blockmatic-w-retro/audits/1

BibTeX 引用

@techreport{blockmatic-blockmatic-w-retro-2026,
  author = {blockmatic},
  title = {w-retro security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/blockmatic-w-retro/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "w-retro security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "blockmatic"
date-released: "2026-09-20"
url: "https://skillstore.io/skills/blockmatic-w-retro/audits/1"
identifiers:
  - type: other
    value: "skillstore:blockmatic-w-retro:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 评分

评分依据证据置信度: 中

架构

可维护性

内容

社区

规范符合性

你能构建什么

Close a development task

Review the request, shipped changes, checks, and documentation before closing a coding task.

Improve team delivery

Capture specific process gaps after a sprint item, project milestone, or cross-functional deliverable.

Evaluate an AI-assisted workflow

Compare the requested work with the AI-produced result and record better checks for the next session.

试试这些提示

Basic task retrospective

Review the completed task in this conversation. State what was requested, what shipped, and one improvement for next time.

Check process gaps

Run a retrospective on this completed task. Identify skipped checks, unclear requirements, extra scope, and missing documentation. Suggest no more than three specific changes.

Compare plan and outcome

Compare the original request, plan, implementation, and validation evidence. Explain the most important differences and recommend up to three changes for the next similar task.

Audit an AI-assisted delivery

Perform a task-specific retrospective of this AI-assisted delivery. Separate outcome gaps from process gaps, cite the available evidence, and prioritize three changes that would improve the next session.

最佳实践

Run the skill after checks are complete and the delivered result is known.
Provide the original request, completed work, validation evidence, and relevant constraints.
Prefer specific recommendations tied to this task over broad process advice.

避免

Do not use the retrospective as a substitute for implementing a required fix.
Do not request generic lessons without sharing task-specific evidence.
Do not turn a short retrospective into an unfocused list of every possible improvement.

常见问题

When should I use this skill?

Use it after completing a task and recording the checks or evidence that support the result.

What does the skill review?

It compares the request with the shipped result and checks for process or evidence gaps.

How many improvements does it suggest?

It suggests at most three concrete changes for the next similar task.

Can it change files or create commits?

No. It responds in chat and does not implement changes, commit files, or open pull requests.

What context produces the best result?

Include the request, plan, completed work, validation checks, constraints, and known uncertainties.

Does it replace a formal incident review?

No. It supports routine task learning and does not replace incident, compliance, or security review processes.

开发者详情

作者

blockmatic

许可证

MIT

Skillstore 修订

版本提示

作者未声明版本。

仓库

https://github.com/blockmatic/basilic-skills/tree/5fa16493413923e31f1df31f7260224f0f291d9e/skills/w-retro

引用

7cd60e5c44d39477abb7a2c00ae0de2cf2a01392

维护新鲜度

2026/9/20

使用量

0 次下载 · 0 次浏览

文件结构

📄 SKILL.md