rewrite

Coordinates cross-store correction of outdated patterns in vault, memory, and episodic history with user approval.

11|1|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/robinslange/learning-loop --skill rewrite-robinslange
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rewrite
Source: https://github.com/robinslange/learning-loop/tree/main/skills/rewrite
Command: npx skills add https://github.com/robinslange/learning-loop --skill rewrite-robinslange

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? Cross-store correction for retracted or updated beliefs across the vault, auto-memory, and episodic history, ensuring consistency and traceability.

## Core Features & Use Cases

  • Cross-store belief rewrites across vault notes, memory, and episodic history.
  • Supersession recording to annotate future retrievals.
  • Phase-based workflow: framing, impact analysis, triage, and execution with user approval.
  • Transition notes for provenance and auditability.

### Quick Start Use /learning-loop:rewrite "old pattern" "new pattern" [reason] to apply a cross-store correction.

Frequently Asked Questions about rewrite

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I update outdated beliefs across my notes and memory stores?

To update outdated beliefs across notes and memory stores, identify the old and new patterns to trigger a cross-store correction. The system generates an impact map tracing primary notes and proposes edits for user approval before applying changes and recording a supersession.

What is the best way to ensure traceability when retracting knowledge from episodic history?

Traceability during knowledge retraction is ensured by recording a supersession after applying cross-store corrections. This annotation guarantees that future episodic searches surface the correction, maintaining provenance and auditability through transition notes.

How does cross-store correction handle vault notes and auto-memory updates?

Cross-store correction handles vault notes and auto-memory updates by tracing primary notes impacted by the belief change. It proposes edits, amends, or archives across these stores and presents an impact map for explicit user approval before executing any modifications.

Can I review proposed memory corrections before they are applied?

Yes, you can review proposed memory corrections before execution. The workflow operates in phases, presenting an impact map of proposed edits, amends, or archives for your explicit approval to ensure alignment before any changes are applied to the stores.

When do I need to record a supersession for knowledge management?

You need to record a supersession for knowledge management immediately after applying a cross-store correction to a retracted or updated belief. This guarantees that future episodic searches surface the correction and maintain consistency across your memory and notes.

Why does my episodic search still return old patterns after a vault edit?

Episodic searches return old patterns after a vault edit because a supersession has not been recorded. Applying cross-store corrections and recording a supersession guarantees that future episodic searches surface the correction and maintain aligned beliefs.