skills.rest
Shiien
ShiienCOMMUNITY
Explore all skills by this author →
GitHub

memory-forget

Remove or archive low-value memories and update MEMORY.md.

7|1|Updated Mar 19, 2026
Search Tags:#decisions#archival#memory#budget#pruning#episodes
One-click install
npx skills add https://github.com/Shiien/Self-Evolved-Research-Framework --skill memory-forget
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-forget
Source: https://github.com/Shiien/Self-Evolved-Research-Framework/tree/main/skills/memory-forget
Command: npx skills add https://github.com/Shiien/Self-Evolved-Research-Framework --skill memory-forget

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delete or archive low-value memories (unretrieved episodes with importance < 5, consolidated episodes with importance < 7, superseded decisions) while preserving architectural decisions, key findings, and active hypotheses. Runs after memory-consolidate if still over the 200-line budget.

Core Features & Use Cases

  • Identify candidates: episodes with no retrievals and importance < 5; consolidated episodes with importance < 7; and superseded decisions.
  • Never forget: architectural decisions, key findings, and active hypotheses.
  • Delete or archive candidates and update MEMORY.md to reflect removals.

Quick Start

Run memory-forget after memory-consolidate when over capacity to prune low-value memories while preserving architectural decisions, key findings, and active hypotheses.

Frequently Asked Questions about memory-forget

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

FAQPage Schema
How do I prune low-value memories to free up memory budget?▼

Remove or archive low-importance episodes and superseded decisions to free up memory budget, preserving architectural decisions, key findings, and active hypotheses while updating the MEMORY.md index.

What memories are safe to delete when consolidating memory files?▼

Unretrieved episodes with importance < 5, consolidated episodes with importance < 7, and superseded decisions are safe to delete or archive. Architectural decisions, key findings, and active hypotheses are always preserved.

When should I run memory pruning on my memory index?▼

Run memory pruning after memory-consolidate if still over the 200-line budget capacity. This targets low-value entries for removal while updating the MEMORY.md index to reflect the changes.

Does memory pruning preserve active hypotheses and architectural decisions?▼

Yes, memory pruning never forgets architectural decisions, key findings, and active hypotheses. It only removes or archives low-importance episodes and superseded decisions while updating the MEMORY.md index.

What's the best way to archive superseded decisions in memory files?▼

Run memory pruning after memory-consolidate to archive superseded decisions and low-importance episodes, freeing up the 200-line memory budget while updating the MEMORY.md index to reflect all removals.

Related Skills

Explore More Data & Analytics Skills→
ShiienShiien
7

memory-consolidate

Cluster unconsolidated episodes into topics and procedures by tag overlap.

Community
Advanced
rockdongrockdong

cognitive-memory

Route user-triggered memory content to core, episodic, semantic, procedural, and vault stores.

Community
Advanced
ShiienShiien
7

memory-write

Persist insights and decisions to memory episodes with deduplication and indexing.

Community
Intermediate
adamrdrewadamrdrew
1

read-memory

Parse memory.md key-value entries and return specific or all keys.

Community
Intermediate