/retro
Post-work command that extracts learnings and updates the learning registry
$ npx -y skills add DanielPodolsky/ownyourcode --agent claude-codeHow it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/retro
Context preview
What this command does when you run it.
Post-work command that extracts learnings and updates the learning registry
Command definition
retro.mdname: retro
description: Post-work command that extracts learnings and updates the learning registry
allowed-tools: Read, Glob, Grep, Write, Edit, AskUserQuestion, mcp__octocode__githubSearchCode, mcp__octocode__githubGetFileContent
/own:retro
> ⚠️ **PLAN MODE WARNING:** Toggle plan mode off before running this command (`shift+tab`). OwnYourCode commands don't work correctly with plan mode.
Extract learnings from completed work and persist them for future reference.
Overview
This command is run **after** completing work (typically after `/own:done`). It: 1. Reflects on what was learned 2. Documents patterns worth reusing 3. Records failures to avoid 4. Updates the learning registry 5. Generates skill files for significant learnings
> "The junior who documents their failures outgrows the senior who repeats them."
---
Execution Flow
Phase 1: What Did You Just Complete?
> "What did you just finish working on?"
If they just ran `/own:done`, reference that context.
Get:
- Feature/task name
- Domain (auth, forms, API, etc.)
- Rough summary
---
Phase 2: Learning Extraction
Ask these questions to extract learnings:
Question 1: The Hard Part
> "What was the hardest part of this work?"
Looking for:
- Technical challenges overcome
- Confusing concepts clarified
- Time-consuming debugging
Question 2: The Insight
> "What do you know now that you didn't know before you started?"
Looking for:
- New understanding of a concept
- Better approach discovered
- "Aha!" moments
Question 3: The Pattern (if applicable)
> "Did you discover or use a pattern that worked well? Something you'd want to reuse?"
Looking for:
- Code patterns worth extracting
- Approaches that solved problems elegantly
- Techniques to remember
Question 4: The Failure (most valuable)
> "Did you make any mistakes or hit any walls? What went wrong before it went right?"
Looking for:
- Bugs that took time to find
- Wrong approaches tried first
- Misunderstandings that caused problems
Question 5: The Advice
> "If you could go back and tell yourself one thing before starting this task, what would it be?"
This often captures the most actionable learning.
---
Phase 3: Categorize Learnings
Based on their answers, categorize:
**Patterns** (reusable solutions):
- Code patterns that worked well
- Approaches worth reusing
- Techniques to remember
**Failures** (anti-patterns):
- Mistakes to avoid
- Wrong approaches
- Time-wasters to skip
**Insights** (general learnings):
- Conceptual understanding
- "Now I get it" moments
- Connections made
---
Phase 4: Generate Learning Artifacts
**All learnings are stored GLOBALLY at `~/ownyourcode/learning/` to persist across projects.**
For Significant Patterns
Create a pattern file at `~/ownyourcode/learning/patterns/[PatternName].md`:
---
name: [Pattern Name]
description: [When to use this pattern]
---
# [Pattern Name]
**Learned:** 2026-01-01
**Domain:** [Domain]
**Context:** [Brief context of when this was learned]
## When to Use
[Situations where this pattern applies]
## The Pattern
[Code example or description]
## Why It Works
[Explanation of why this is effective]
## Watch Out For
[Common mistakes or edge cases]
## Related
- [Link to original task if applicable]
For Significant Failures
Create a failure doc at `~/ownyourcode/learning/failures/[Topic].md`:
# FAILURE: [Short Description]
**Date:** 2026-01-01
**Domain:** [Domain]
**Time Lost:** [Rough estimate]
## What Happened
[Description of the failure]
## Root Cause
[Why it happened]
## The Fix
[How it was resolved]
## How to Avoid
[What to do differently next time]
## Red Flags
[Warning signs to watch for]
---
Phase 5: Update Global Learning Registry
Append to `~/ownyourcode/learning/LEARNING_REGISTRY.md`:
### 2026-01-01: [Title]
**Domain:** [Domain]
**Type:** Pattern | Failure | Insight
**Summary:** [One-line summary]
**Key Insight:** [The main takeaway]
**Location:** [Path to skill file or failure doc if created]
Also update the index tables at the top of the registry.
---
Phase 6: Competency Check
Based on accumulated learnings, assess growth:
┌─────────────────────────────────────────┐
│ COMPETENCY SNAPSHOT │
├─────────────────────────────────────────┤
│ │
│ Total Patterns Documented: [X] │
│ Total Failures Documented: [Y] │
│ Total Insights Captured: [Z] │
│ │
│ Domains Explored: │
│ • Auth: ████████░░ (8 learnings) │
│ • Forms: ████░░░░░░ (4 learnings) │
│ • API: ██░░░░░░░░ (2 learnings) │
│ │
│ Current Level: ⭐⭐⭐ (Intermediate) │
│ │
│ "You're building a solid foundation. │
│ Keep documenting — this compounds." │
│ │
└─────────────────────────────────────────┘
---
Phase 7: Summary
┌─────────────────────────────────────────┐
│ RETROSPECTIVE COMPLETE │
├─────────────────────────────────────────┤
│ │
│ Task: [Feature Name] │
│ Domain: [Domain] │
│ │
│ LEARNINGS CAPTURED │
│ ───────────────── │
│ ✅ Pattern: [Name] → saved to skills │
│ ⚠️ Failure: [Name] → documented │
│ 💡 Insight: "[Key insight]" │
│ │
│ GLOBAL REGISTRY UPDATED │
│ ───────────────── │
│ ~/ownyourcode/learning/LEARNING_REGISTRY│
│ │
│ NEXT TIME │
│ ───────────────── │
│ Run /own:advise before your next task │
│
Read more
name: retro description: Post-work command that extracts learnings and updates the learning registry allowed-tools: Read, Glob, Grep, Write, Edit, AskUserQuestion, mcp__octocode__githubSearchCode, mcp__octocode__githubGetFileContent
/own:retro
> ⚠️ **PLAN MODE WARNING:** Toggle plan mode off before running this command (`shift+tab`). OwnYourCode commands don't work correctly with plan mode.
Extract learnings from completed work and persist them for future reference.
Overview
This command is run **after** completing work (typically after `/own:done`). It: 1. Reflects on what was learned 2. Documents patterns worth reusing 3. Records failures to avoid 4. Updates the learning registry 5. Generates skill files for significant learnings
> "The junior who documents their failures outgrows the senior who repeats them."
---
Execution Flow
Phase 1: What Did You Just Complete?
> "What did you just finish working on?"
If they just ran `/own:done`, reference that context.
Get:
- Feature/task name
- Domain (auth, forms, API, etc.)
- Rough summary
---
Phase 2: Learning Extraction
Ask these questions to extract learnings:
Question 1: The Hard Part
> "What was the hardest part of this work?"
Looking for:
- Technical challenges overcome
- Confusing concepts clarified
- Time-consuming debugging
Question 2: The Insight
> "What do you know now that you didn't know before you started?"
Looking for:
- New understanding of a concept
- Better approach discovered
- "Aha!" moments
Question 3: The Pattern (if applicable)
> "Did you discover or use a pattern that worked well? Something you'd want to reuse?"
Looking for:
- Code patterns worth extracting
- Approaches that solved problems elegantly
- Techniques to remember
Question 4: The Failure (most valuable)
> "Did you make any mistakes or hit any walls? What went wrong before it went right?"
Looking for:
- Bugs that took time to find
- Wrong approaches tried first
- Misunderstandings that caused problems
Question 5: The Advice
> "If you could go back and tell yourself one thing before starting this task, what would it be?"
This often captures the most actionable learning.
---
Phase 3: Categorize Learnings
Based on their answers, categorize:
**Patterns** (reusable solutions):
- Code patterns that worked well
- Approaches worth reusing
- Techniques to remember
**Failures** (anti-patterns):
- Mistakes to avoid
- Wrong approaches
- Time-wasters to skip
**Insights** (general learnings):
- Conceptual understanding
- "Now I get it" moments
- Connections made
---
Phase 4: Generate Learning Artifacts
**All learnings are stored GLOBALLY at `~/ownyourcode/learning/` to persist across projects.**
For Significant Patterns
Create a pattern file at `~/ownyourcode/learning/patterns/[PatternName].md`:
--- name: [Pattern Name] description: [When to use this pattern] --- # [Pattern Name] **Learned:** 2026-01-01 **Domain:** [Domain] **Context:** [Brief context of when this was learned] ## When to Use [Situations where this pattern applies] ## The Pattern [Code example or description] ## Why It Works [Explanation of why this is effective] ## Watch Out For [Common mistakes or edge cases] ## Related - [Link to original task if applicable]
For Significant Failures
Create a failure doc at `~/ownyourcode/learning/failures/[Topic].md`:
# FAILURE: [Short Description] **Date:** 2026-01-01 **Domain:** [Domain] **Time Lost:** [Rough estimate] ## What Happened [Description of the failure] ## Root Cause [Why it happened] ## The Fix [How it was resolved] ## How to Avoid [What to do differently next time] ## Red Flags [Warning signs to watch for]
---
Phase 5: Update Global Learning Registry
Append to `~/ownyourcode/learning/LEARNING_REGISTRY.md`:
### 2026-01-01: [Title] **Domain:** [Domain] **Type:** Pattern | Failure | Insight **Summary:** [One-line summary] **Key Insight:** [The main takeaway] **Location:** [Path to skill file or failure doc if created]
Also update the index tables at the top of the registry.
---
Phase 6: Competency Check
Based on accumulated learnings, assess growth:
┌─────────────────────────────────────────┐ │ COMPETENCY SNAPSHOT │ ├─────────────────────────────────────────┤ │ │ │ Total Patterns Documented: [X] │ │ Total Failures Documented: [Y] │ │ Total Insights Captured: [Z] │ │ │ │ Domains Explored: │ │ • Auth: ████████░░ (8 learnings) │ │ • Forms: ████░░░░░░ (4 learnings) │ │ • API: ██░░░░░░░░ (2 learnings) │ │ │ │ Current Level: ⭐⭐⭐ (Intermediate) │ │ │ │ "You're building a solid foundation. │ │ Keep documenting — this compounds." │ │ │ └─────────────────────────────────────────┘
---
Phase 7: Summary
┌─────────────────────────────────────────┐ │ RETROSPECTIVE COMPLETE │ ├─────────────────────────────────────────┤ │ │ │ Task: [Feature Name] │ │ Domain: [Domain] │ │ │ │ LEARNINGS CAPTURED │ │ ───────────────── │ │ ✅ Pattern: [Name] → saved to skills │ │ ⚠️ Failure: [Name] → documented │ │ 💡 Insight: "[Key insight]" │ │ │ │ GLOBAL REGISTRY UPDATED │ │ ───────────────── │ │ ~/ownyourcode/learning/LEARNING_REGISTRY│ │ │ │ NEXT TIME │ │ ───────────────── │ │ Run /own:advise before your next task │ │
Claude Code workflow for AI-mentored development. Work efficiently with Spec-Driven Development and the 6 Gates. Built to fight cognitive offloading — for developers using AI to grow and maintain ownership.
Repo: DanielPodolsky/ownyourcode
Other commands on ownyourcode.
- /advise
Pre-work command that queries past learnings and leverages MCPs before starting a new task
Open command - /docs
Guide the junior through writing documentation for their project or code
Open command - /done
Complete a task with 6 Gates verification, code review, and career value extraction
Open command - /feature
Create a feature specification using spec-driven development
Open command - /guide
Get implementation guidance for the current task
Open command - /init
Initialize OwnYourCode project with mission, stack, and roadmap
Open command

