memory-review

Name: memory-review
Availability: InStock
Author: ONEBRAIN

Audit and prune persisted memory entries by reviewing MEMORY-INDEX.md files.

23|2|Updated Mar 20, 2026

Search Tags:#audit #memory #archiving #confidence #deprecate #knowledge vault #index sync

Ready to run

npx skills add https://github.com/onebrain-ai/onebrain --skill memory-review-onebrain-ai

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: memory-review
Source: https://github.com/onebrain-ai/onebrain/tree/main/.claude/plugins/onebrain/skills/memory-review
Command: npx skills add https://github.com/onebrain-ai/onebrain --skill memory-review-onebrain-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OneBrain memory can accumulate outdated or low-confidence entries, making future sessions less accurate and harder to trust; this Skill helps you review everything and keep only what still matters.

Core Features & Use Cases

Interactive, entry-by-entry pruning: keep, update confidence/metadata, flag for later review, deprecate, or delete entries with clear per-entry status.
Safe update workflow: updates are staged through a guided two-step process and only committed after confirmation.
Index consistency maintenance: deprecations and deletions update MEMORY-INDEX.md counters and metadata so the vault stays coherent.

Quick Start

When you think your OneBrain knowledge is getting noisy, ask an AI to run memory-review so it can walk through every memory file and let you keep, update, deprecate, or delete each one interactively.

Frequently Asked Questions about memory-review

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

FAQPage Schema

How do I audit and clean up outdated AI memory entries?▼

To audit memory entries, you can run an interactive review process that walks through stored memory files one by one, letting you keep, update confidence, deprecate, or delete each entry safely.

What is the best way to prune accumulated knowledge vault entries?▼

Pruning a knowledge vault involves reviewing each memory file individually, recalibrating confidence metadata, and archiving or removing outdated entries to keep the vault coherent for future sessions.

Can I update memory index counters automatically when deprecating entries?▼

Deprecating entries automatically updates MEMORY-INDEX.md counters and metadata, ensuring the knowledge vault stays coherent and accurately reflects the active memory status.

How does the confidence recalibration process work for stored memory?▼

Confidence recalibration works by staging updates through a guided two-step process, allowing you to adjust per-entry metadata and only committing the changes after explicit user confirmation.

When should I review and deprecate active memory files?▼

You should review and deprecate active memory files when accumulated knowledge becomes noisy or outdated, making future AI sessions less accurate and harder to trust.

Do I need explicit confirmation to delete memory entries?▼

Deleting memory entries requires explicit user confirmation, as all updates are staged through a guided two-step process and committed only after you approve the changes.