
lab-retro
Final retrospective and self-assessment for participants of Claude Code Lab. Runs four sequential interactive parts — progress audit, best prompt, monthly plan, and feedback — using AskUserQuestion. Triggers on "/lab-retro", "lab retrospective", "claude code lab final", or after completing the 6-week Claude Code Lab cohort.
- Installs
- 177
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/glebis/claude-skills --skill lab-retro- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads lab-retro automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Claude Code Lab — Final Retrospective
This skill walks a Claude Code Lab graduate through four sequential exercises that consolidate their learning, capture their best work, plan next steps, and collect structured feedback for the organizer.
How to run
Default flow: run all four parts in order. The user can also jump to a specific part with /lab-retro 2 (or just say "part 3").
Between parts, briefly summarize what just happened and ask "ready for part N?" so the user controls the pace.
All artifacts are saved into a single folder lab-retro-output/ in the current working directory:
01-progress.md02-best-prompt.md03-month-plan.md04-feedback.json+04-feedback-report.md
Create the folder if missing.
Part 1 — Progress audit
Goal: help the participant see concrete before/after.
Use AskUserQuestion:
-
"Что вы умели ДО лаборатории?" (multiSelect)
- Работал с ChatGPT/Claude в браузере
- Использовал CLI инструменты
- Писал код
- Автоматизировал задачи
- Работал с API
-
"Что вы умеете ПОСЛЕ?" (multiSelect)
- Работаю с Claude Code ежедневно
- Настроил MCP-серверы
- Создал свой Skill
- Автоматизировал реальную задачу
- Задеплоил что-то в веб
- Использую субагентов
- Пишу и читаю CLAUDE.md осознанно
-
"Сколько часов в неделю экономит Claude Code?" (singleSelect)
- <2
- 2–5
- 5–10
- 10+
Then output a markdown table comparing before/after with skill levels (0–5) and save to lab-retro-output/01-progress.md.
Part 2 — Best prompt
Goal: turn one prompt the participant is proud of into a reusable Skill.
Ask the participant: "Скопируйте или опишите ваш самый полезный промт из лабы."
Then AskUserQuestion:
-
"Для какой задачи был промт?" (singleSelect)
- Автоматизация рутины
- Создание контента/документации
- Анализ данных/исследование
- Прототипирование/разработка
- Личный workflow / PKM
-
"Что сделало его эффективным?" (multiSelect)
- Хороший контекст в CLAUDE.md
- Чёткие критерии успеха
- Разбиение на шаги
- Использование Skills/MCP
- Примеры в промте
- Ограничения и анти-критерии
Reformat the prompt as a proper Skill (frontmatter + body), suggest an description line that would trigger it, and save to lab-retro-output/02-best-prompt.md. Suggest where to put it (~/.claude/skills/<name>/SKILL.md).
Part 3 — Month plan
Goal: concrete 4-week plan so momentum doesn't die after the cohort.
AskUserQuestion:
-
"Главная рабочая задача на месяц?" (singleSelect)
- Проект для клиента/работодателя
- Свой продукт/стартап
- Автоматизация текущих процессов
- Обучение и развитие навыков
-
"Сколько времени в неделю готовы уделять?" (singleSelect)
- 15–30 минут
- 1–2 часа
- 3–5 часов
- Каждый день
-
"Какой риск выгорания вы оцениваете для себя?" (singleSelect)
- Низкий — у меня устойчивый ритм
- Средний — иногда залипаю
- Высокий — уже узнал себя в красных флагах
If risk = high, insert a mandatory rest day into the plan and a recommendation to read the AI hygiene slide again.
Generate lab-retro-output/03-month-plan.md with:
- 4 weekly goals
- 1–3 concrete prompts per week
- Success criterion per week
- Stop-conditions (when to pause)
Part 4 — Feedback for the organizer
Goal: structured feedback that goes back to the lab organizer.
AskUserQuestion:
- "Оцените лабу в целом (NPS)" (singleSelect: 0–10)
- "Самая ценная встреча?" (singleSelect: M01 / M03 / M05 / M07 / M09 / M11)
- "Самая ценная тема за все 6 недель?" (multiSelect)
- Основы Claude Code
- Промтинг и контекст
- Архитектура и субагенты
- MCP / Skills / Hooks
- Agent SDK и деплой
- Evals и качество
- AI-гигиена
- "Что улучшить?" (free text)
- "Главное препятствие, с которым вы столкнулись?" (free text)
- "Согласны ли поделиться отзывом публично?" (singleSelect: да / да-анонимно / нет)
Save TWO files:
lab-retro-output/04-feedback.json— structured for the organizerlab-retro-output/04-feedback-report.md— human-readable summary for the participant
Then submit to the public proxy (no secrets needed):
curl -sS -X POST https://lab-feedback-proxy.vercel.app/api/feedback \
-H "Content-Type: application/json" \
-d "$(jq -nc --arg name "<participant name>" --slurpfile notes lab-retro-output/04-feedback.json '{name:$name, notes:($notes[0]|tostring)}')"
The proxy forwards to Baserow table 746002 with a server-side token. Response is {"ok":true,"row_id":<N>}. Confirm row ID with the participant.
If the request fails, fall back to local files only and tell the participant: "submit failed — your feedback is saved locally in lab-retro-output/04-feedback.json, send it to the organizer manually."
Closing
After all four parts, print:
- Where the four files live
- One sentence: "Ваш фреймворк — LOOP. Ваш план — в файле 03. Ваш следующий шаг — открыть его в понедельник утром."
- Suggest committing the
lab-retro-output/folder to a personal repo or vault.
Constraints
- Always use
AskUserQuestionfor structured questions — don't ask in plain text - Never skip a part silently; if the user opts out, write a one-line stub in the corresponding file
- Russian by default (audience is Russian-speaking); switch to English if the user replies in English
- Don't be sycophantic in the feedback report — surface honest patterns
FAQ
What does the lab-retro skill do?
Final retrospective and self-assessment for participants of Claude Code Lab. Runs four sequential interactive parts — progress audit, best prompt, monthly plan, and feedback — using AskUserQuestion. Triggers on "/lab-retro", "lab retrospective", "claude code lab final", or after completing the 6-week Claude Code Lab cohort.
How do I install lab-retro?
Run npx skills add https://github.com/glebis/claude-skills --skill lab-retro in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/lab-retro/ (personal) or .claude/skills/lab-retro/ (project).
Which AI agents can use lab-retro?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is lab-retro safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains lab-retro?
It is published by glebis in glebis/claude-skills. It has 177 recorded installs.
More from glebis/claude-skills
google-image-search
glebis/claude-skills
Search and download images via Google Custom Search API with LLM-powered selection. This skill should be used when finding images for articles, presentations, research documents, or enriching Obsidian notes with relevant visuals. Supports simple queries, batch processing from JSON config, automatic config generation from terms, and full note enrichment with automatic image insertion below headings.
pdf-generation
glebis/claude-skills
Professional PDF generation from markdown using Pandoc with Eisvogel template and EB Garamond fonts. Use when converting markdown to PDF, creating white papers, research documents, marketing materials, or technical documentation. Supports both English and Russian documents with professional typography and color-coded themes. Mobile-optimized layout (6x9) by default for Telegram bot context, desktop/print layout (A4) for other contexts.
health-data
glebis/claude-skills
Open the skill to read its SKILL.md instructions.
firecrawl-research
glebis/claude-skills
Open the skill to read its SKILL.md instructions.
deep-research
glebis/claude-skills
This skill should be used when conducting comprehensive research on any topic using the OpenAI Deep Research API. It automates prompt enhancement through interactive clarifying questions, saves research parameters, and executes deep research with web search capabilities. Use when the user asks for in-depth analysis, investigation, research summaries, or topic exploration.
presentation-generator
glebis/claude-skills
Generate interactive HTML presentations with neobrutalism styling, ASCII art decorations, and Agency brand colors. Outputs HTML (interactive with navigation), PNG (individual slides via Playwright), and PDF. References brand-agency skill for colors and typography. Use when creating presentations, slide decks, pitch materials, or visual summaries.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available