Capture learnings after tasks. Triggers on task completion, repeated mistakes, retrospective requests, or "what did I learn".
After completing a task, briefly reflect and capture learnings.
Ask yourself:
| Learning Type | Action |
|---|---|
| Pattern/gotcha | Store in Memory MCP |
| Missing automation | Create issue/TODO |
| Repeated mistake | Update .claude/rules/ |
| Complex process | Create .claude/skills/ |
When storing a learning in Memory MCP:
mcp__memory__create_entities({
entities: [{
name: "Learning_TOPIC",
entityType: "development_learning",
observations: [
"Context: What was happening",
"Learning: What was learned",
"Application: How to apply it"
]
}]
})
| Issue | Solution |
|---|---|
| Same mistake twice | Add rule to .claude/rules/ |
| Manual process > 5 min | Create skill or hook |
| New tool/pattern | Store in Memory MCP |
| Missing documentation | Update docs |
When you make the same mistake twice:
# .claude/rules/my-rules.md
---