memory-manager

Persist validated niches and learnings to JSON memory files across sessions.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/thebird-create/MOM --skill memory-manager-thebird-create
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-manager
Source: https://github.com/thebird-create/MOM/tree/main/.claude/skills/memory-manager
Command: npx skills add https://github.com/thebird-create/MOM --skill memory-manager-thebird-create

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents the loss of valuable insights and validated business decisions by automatically persisting them across AI sessions. This ensures that past learnings consistently inform future actions, eliminating the need for manual tracking and recall.

Core Features & Use Cases

  • Automated Persistence: Automatically saves "GO" decisions from niche validation and archives "SKIP" decisions for future pattern spotting.
  • Deterministic Retention: Applies strict rules for data cleanup, archiving entries older than 90 days and retaining the top 100 highest-confidence items per brand.
  • Query Guidance: Provides clear command-line examples (e.g., jq, grep) for easily accessing and analyzing stored information.
  • Use Case: When the pod-research skill identifies a promising niche, this Skill automatically logs that decision, along with all relevant data, into a structured memory file, making it instantly recallable for future product development or strategic planning.

Quick Start

Save the learning that "sustainable pet products" is a GO niche with 0.90 confidence, validated on 2025-12-07 for the "LWF" brand.

Frequently Asked Questions about memory-manager

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

FAQPage Schema
How do I automatically save and retain business decisions across AI sessions?

Automated knowledge retention persists validated niche decisions by storing GO determinations and SKIP entries in structured JSON files, enforcing a 90-day archive window and maintaining the top 100 highest-confidence items per brand to ensure past learnings consistently inform future product strategy.

What happens to old entries when I persist business learnings and validated niches?

Memory retention automatically enforces a 90-day archive window, removing entries older than that threshold, while capping stored data at the top 100 highest-confidence items per brand to keep decision logs clean and actionable.

How do I query and access saved niche validation decisions from memory storage?

Stored decisions are saved as JSON at .claude/memories/validated_niches.json and .claude/data/skipped_niches.json, with shell-ready query guidance using tools like jq and grep to extract and analyze retained learnings by niche, brand, or confidence score.

Can I track both successful GO decisions and rejected SKIP decisions in the same memory system?

Yes; the system deterministically logs both GO decisions from validated niches and SKIP entries into separate JSON files, allowing you to spot patterns across accepted and rejected opportunities for consistent strategic planning.

What data do I need to provide when saving a validated niche to memory?

Save learnings by supplying the niche name, confidence score, validation date, and brand identifier; the system deterministically stores this metadata in structured JSON format, enabling instant recall and traceable memory operations for future analysis.