memory-review

Clean and consolidate unreviewed memories in a local database.

51|5|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/m13v/ai-browser-profile --skill memory-review-m13v
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-review
Source: https://github.com/m13v/ai-browser-profile/tree/main/review
Command: npx skills add https://github.com/m13v/ai-browser-profile --skill memory-review-m13v

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Clean up and consolidate unreviewed memories stored in the local memories database to improve data quality and reliability.

Core Features & Use Cases

  • Bulk cleanup: remove noisy, junk, and noise entries in unreviewed memories.
  • Supersession repair: fix corrupted single-value key chains and restore correct values.
  • Per-entry review workflow: categorize remaining entries as KEEP, DELETE, MERGE, or FIX for a clean profile.
  • Periodic maintenance: run after extraction to maintain a high-quality memories dataset.

Quick Start

Run the memory-review workflow to process new memories in ~/ai-browser-profile/memories.db.

Frequently Asked Questions about memory-review

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

FAQPage Schema
How do I clean up unreviewed memories in a local database?

To clean up unreviewed memories in a local database, you can run a bulk cleanup workflow to remove noisy entries and categorize remaining data as KEEP, DELETE, MERGE, or FIX. This improves overall data quality.

What is the best way to fix corrupted single-value key chains in memory datasets?

The best way to fix corrupted single-value key chains is through supersession repair, which restores correct values and resolves data corruption issues across ingestion pipelines.

How do I deduplicate and consolidate memories after data ingestion?

You can deduplicate and consolidate memories after ingestion by running a per-entry review workflow that categorizes entries for KEEP, DELETE, MERGE, or FIX, ensuring a clean profile.

When should I run a memory review workflow to maintain data quality?

You should run a memory review workflow periodically after data extraction to maintain a high-quality memories dataset by locating unreviewed entries and applying rules to fix or delete them.

Can I apply deletion and merging rules to specific entries in a local memories database?

Yes, you can apply deletion and merging rules to specific entries in a local memories database through a per-entry review workflow that records reviewed statuses and handles common data issues.

Why does my memories database have noisy entries and how can I remove them?

Your memories database has noisy entries due to ingestion pipelines, and you can remove them using a bulk cleanup workflow that targets junk and noise data to improve reliability.