One-click install
npx skills add https://github.com/ahostbr/liteharness --skill ls-dream
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ls-dream_
Source: https://github.com/ahostbr/liteharness/tree/main/liteharness/catalog/skills/ls-dream_
Command: npx skills add https://github.com/ahostbr/liteharness --skill ls-dream

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents memory bloat and contradictions by consolidating scattered notes and recent signals into a clean, durable memory set.

Core Features & Use Cases

  • Reflective memory consolidation: Synthesizes recent learnings into well-organized, long-lived memory files.
  • Prune and resolve drift: Updates existing memory topics, removes stale or contradicted facts, and keeps an index that stays readable.
  • Transcript-guided focusing: Greps narrow terms in large JSONL transcripts to gather only the most relevant context before writing.

Quick Start

Ask your agent to run the ls-dream_ memory consolidation pass to orient future sessions by updating MEMORY.md and the top-level memory topic files.

Frequently Asked Questions about ls-dream_

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

FAQPage Schema
How do I consolidate scattered notes into a durable memory index?

Memory consolidation synthesizes recent learnings and prunes drift to create a durable memory index. It updates existing topics, removes stale facts, and resolves contradictions so your knowledge base remains clean and readable without bloat.

What is the best way to prevent memory bloat and contradictions in knowledge management?

Preventing memory bloat requires synthesizing scattered notes into durable memory files and pruning outdated topics. By detecting drift in existing memories and removing near-duplicates during consolidation, you keep a clean knowledge base.

How do I keep a MEMORY.md index under 200 lines during memory consolidation?

Keeping a MEMORY.md index under 200 lines requires pruning stale facts and merging updates without creating near-duplicates. Memory consolidation resolves contradictions and synthesizes topics to keep the index readable and strictly bounded.

How do I search large JSONL transcript files for targeted context?

Searching large JSONL transcripts involves grepping narrow terms to gather only relevant context. This transcript-guided focusing extracts targeted information before writing updates to your memory files.

When do I need memory pruning and drift detection for my logs?

Memory pruning and drift detection are needed during reflective maintenance sessions when reviewing daily logs. If existing memories contain stale facts, pruning resolves the drift and updates topics to maintain a consistent memory set.

Does memory consolidation require existing memory files to work?

Memory consolidation requires reading MEMORY.md for the current index to merge updates correctly. It uses this existing index to avoid near-duplicate creation, convert relative dates to absolute dates, and organize durable memory files effectively.