self-improve

Self-Improvement Loop — Core Skill

AuthorAgent gets better every time it works. This skill creates a persistent learning loop where the agent tracks what works, what fails, and what the user prefers — then applies those lessons to future tasks.

How It Works

The Loop

  ┌─────────────────────────────┐
  │                             │
  │    1. DO THE WORK           │
  │    (goal step, writing,     │
  │     research, etc.)         │
  │                             │
  └──────────┬──────────────────┘
             │
             ▼
  ┌─────────────────────────────┐
  │                             │
  │    2. OBSERVE RESULT        │
  │    Did the user accept it?  │
  │    Did they revise it?      │
  │    Did it trigger an error? │
  │    How long did it take?    │
  │                             │
  └──────────┬──────────────────┘
             │
             ▼
  ┌─────────────────────────────┐
  │                             │
  │    3. EXTRACT LESSON        │
  │    What specifically        │
  │    went right or wrong?     │
  │    What pattern emerges?    │
  │                             │
  └──────────┬──────────────────┘
             │
             ▼
  ┌─────────────────────────────┐
  │                             │
  │    4. STORE LESSON          │
  │    Write to learning log    │
  │    (workspace/memory/       │
  │     improvement-log.jsonl)  │
  │                             │
  └──────────┬──────────────────┘
             │
             ▼
  ┌─────────────────────────────┐
  │                             │
  │    5. APPLY LESSONS         │
  │    Before each new task,    │
  │    check the log for        │
  │    relevant lessons and     │
  │    adjust behavior          │
  │                             │
  └──────────┬──────────────────┘
             │
             └──────── back to step 1

What Gets Tracked

Every lesson entry in improvement-log.jsonl contains:

{
  "id": "lesson-042",
  "timestamp": "2026-02-24T14:30:00Z",
  "category": "writing",
  "trigger": "user_revision",
  "context": "Chapter 3 of thriller project",
  "observation": "User rewrote all dialogue tags from creative tags to simple said/asked",
  "lesson": "This user strongly prefers invisible dialogue tags (said/asked). Do not use creative tags like 'exclaimed', 'muttered', 'hissed' unless the user specifically asks.",
  "confidence": 0.9,
  "applied_count": 0,
  "source": "user_feedback"
}