How do I re-enable learning?

Run /learn-on to resume automatic extraction and query counting.

Are existing insights deleted when learning is turned off?

No. All captured insights remain; only automatic extraction and query counting stop.

Will manual learning (/learn) still modify the knowledge base?

Yes. Manual extraction remains available to curate or add insights.

learn-off

npx machina-cli add skill aiskillstore/marketplace/learn-off --openclaw

Files (1)

SKILL.md

1.1 KB

Learn Off

Disable continuous learning mode. Automatic insight extraction will stop.

What This Does

Deactivates continuous learning mode:

Automatic extraction stops
Query counting stops
Manual /learn commands still work

Instructions

Read knowledge/state.json

Update the state:

{
  "learning_mode": false,
  "learning_mode_since": null
}

Write updated state back to knowledge/state.json
Confirm to user with summary of what was learned

Output Format

Continuous Learning: DISABLED
─────────────────────────────
Learning mode is now inactive.

Session summary:
  - Extractions performed: X
  - Queries analyzed: Y
  - Insights captured: Z

Manual extraction is still available via /learn.
Use /knowledge to view accumulated insights.

Notes

Disabling learning mode does not delete any captured insights
The knowledge base remains available for reference
You can re-enable with /learn-on at any time

Source

git clone https://github.com/aiskillstore/marketplace/blob/main/skills/0xrdan/learn-off/SKILL.mdView on GitHub

Overview

Learn-off pauses automatic insight extraction and query counting while preserving existing insights. Manual /learn commands remain available for targeted updates and review.

How This Skill Works

It reads knowledge/state.json, sets learning_mode to false and learning_mode_since to null, writes the updated state back, and reports a DISABLED status. Automatic extraction and query counting stop, but manual learning via /learn remains functional.

When to Use It

When you need to pause automatic data extraction for audits, privacy reviews, or troubleshooting.
Before large-scale data updates to prevent drift in the knowledge base.
During troubleshooting to isolate learning-related issues without new extractions.
In offline or low-resource environments to reduce processing load.
When you want to review accumulated insights using /knowledge before re-enabling learning.

Quick Start

Step 1: Read knowledge/state.json
Step 2: Update to { "learning_mode": false, "learning_mode_since": null }
Step 3: Write back and verify the status shows Continuous Learning: DISABLED

Best Practices

Verify the state change by inspecting knowledge/state.json after updating.
Use the provided DISABLED output to confirm the mode is off.
Document what was learned prior to turning off learning for future reference.
Remember to re-enable with /learn-on when ready.
Rely on manual /learn for selective updates while learning is off.

Example Use Cases

Audit compliance: pause auto-learning before a formal review.
Production stability: disable learning during a deployment window.
Data cleanup: pause auto-extractions while reorganizing the KB.
Privacy-sensitive projects: turn off learning to minimize data processing.
Troubleshooting: reproduce issues without automatic insights changing the dataset.

Frequently Asked Questions

Add this skill to your agents