Skip to main content
skill CX Codex

w-retro — Claude Code skill for task retrospectives

The w-retro Claude Code skill reviews completed work and suggests changes for the next similar task. It runs after a task is finished.

MIT 647 installs ★ 1 updated 2026-10-02

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add blockmatic-icebox/basilic-skills --skill w-retro -a claude-code
Open on GitHub

What w-retro does

The skill compares the original request with the delivered work, then reviews the plan and checks. It points out process gaps such as skipped checks, unclear requirements, extra scope, or missing documentation.

It suggests up to three concrete changes for the next similar task, grounded in the work just reviewed. The exchange stays in chat. The skill does not implement changes, create commits, or open a PR.

What it helps with

  • compare a request with the delivered result
  • spot skipped checks or unclear requirements
  • choose improvements for the next similar task

Example prompts for the agent

  • Review the finished task and suggest what to change next time.
  • Compare the original request with what shipped and identify gaps in the checks.
  • Look at the plan and documentation for this task, then propose improvements.

How to install w-retro

1Skills CLI (Vercel)

npx skills add blockmatic-icebox/basilic-skills --skill w-retro -a claude-code

2Manually: put the files in place

~/.claude/skills/w-retro/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/w-retro/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about w-retro

How do I install w-retro in Claude Code?

Run npx skills add blockmatic-icebox/basilic-skills --skill w-retro -a claude-code. The skill file is at ~/.claude/skills/w-retro/SKILL.md.

What does w-retro do?

It reviews completed work by comparing the request with the result, identifying process gaps, and proposing up to three specific changes for a similar future task.

Can w-retro change code or open a PR?

No. It responds in chat and does not implement changes, create commits, or open a PR.

Source text

SKILL.md · MIT
---
name: w-retro
description: After finishing work, answer what you would do better next time.
disable-model-invocation: true
---

Look back at the completed task, plan, and checks. Chat only. Do not implement, commit, or open a PR.

1. Restate what was requested versus what shipped.
2. Name process or evidence gaps: skipped checks, unclear requirements, extra scope, missing docs.
3. Suggest at most three concrete changes for the next similar task. Cite this task, not generic advice.

© blockmatic-icebox · MIT · blockmatic-icebox/basilic-skills

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set45

The whole set →

Project sponsors

Companies that keep this analytics open