> ## Documentation Index
> Fetch the complete documentation index at: https://docs.drdj.me/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenClaw Learning Loop

> Controlled observe-synthesize-approve-apply loop for improving agent autonomy safely.

## Goal

Learn manager and operator patterns over time without unsafe self-modification.

## Learning Pipeline

1. Observe:
   * completed tasks
   * approvals/rejections
   * communication edits
2. Synthesize:
   * summarize patterns into operating profile
3. Approve:
   * manager reviews profile changes
4. Apply:
   * update prompts/skill policy only after approval

## Data Files

* Input:
  * `/docs/agents/learning/interaction-log.jsonl`
* Output:
  * `/docs/agents/learning/OPERATING_PROFILE.md`

## Command

```bash theme={null}
pnpm openclaw:learn
```

## Safety Guarantees

1. No autonomous policy mutation.
2. No secrets in learning data.
3. External communication privacy rules remain fixed.
4. Human signoff required before behavior promotion.