# Compound Learnings

Transform session learnings into permanent capabilities (skills, rules, agents).

- Repository: https://github.com/parcadei/Continuous-Claude-v3 (3947 stars)
- Category: AI and agents
- License: MIT
- Last updated: 2026-01-26
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add parcadei/Continuous-Claude-v3 --skill compound-learnings -g -a claude-code -y`
Codex: `npx skills add parcadei/Continuous-Claude-v3 --skill compound-learnings -g -a codex -y`
Cursor: `npx skills add parcadei/Continuous-Claude-v3 --skill compound-learnings -g -a cursor -y`
Gemini CLI: `npx skills add parcadei/Continuous-Claude-v3 --skill compound-learnings -g -a gemini-cli -y`

## Description

Transform session learnings into permanent capabilities (skills, rules, agents). Use when asked to "improve setup", "learn from sessions", "compound learnings", or "what patterns should become skills".

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Compound Learnings

Transform ephemeral session learnings into permanent, compounding capabilities.

## When to Use

- "What should I learn from recent sessions?"
- "Improve my setup based on recent work"
- "Turn learnings into skills/rules"
- "What patterns should become permanent?"
- "Compound my learnings"

## Process

### Step 1: Gather Learnings

```bash
# List learnings (most recent first)
ls -t $CLAUDE_PROJECT_DIR/.claude/cache/learnings/*.md | head -20

# Count total
ls $CLAUDE_PROJECT_DIR/.claude/cache/learnings/*.md | wc -l
```

Read the most recent 5-10 files (or specify a date range).

### Step 2: Extract Patterns (Structured)

For each learnings file, extract entries from these specific sections:

| Section Header | What to Extract |
|----------------|-----------------|
| `## Patterns` or `Reusable techniques` | Direct candidates for rules |
| `**Takeaway:**` or `**Actionable takeaway:**` | Decision heuristics |
| `## What Worked` | Success patterns |
| `## What Failed` | Anti-patterns (invert to rules) |
| `## Key Decisions` | Design principles |

Build a frequency table as you go:

```markdown
| Pattern | Sessions | Category |
|---------|----------|----------|
| "Check artifacts before editing" | abc, def, ghi | debugging |
| "Pass IDs explicitly" | abc, def, ghi, jkl | reliability |
```

### Step 2b: Consolidate Similar Patterns

Before counting, merge patterns that express the same principle:

**Example consolidation:**
- "Artifact-first debugging"
- "Verify hook output by inspecting files"
- "Filesystem-first debugging"
→ All express: **"Observe outputs before editing code"**

Use the most general formulation. Update the frequency table.

### Step 3: Detect Meta-Patterns

**Critical step:** Look at what the learnings cluster around.

If >50% of patterns relate to one topic (e.g., "hooks", "tracing", "async"):
→ That topic may need a **dedicated skill** rather than multiple rules
→ One skill compounds better than five rules

The rest: https://github.com/parcadei/Continuous-Claude-v3/blob/HEAD/.claude/skills/compound-learnings/SKILL.md