ktao732084-archktao732084-archCommunityยท1 Agent Skills Included

openclaw_memory_supersystem-v1.0

Persistent layered memory with decay, retrieval, and consolidation

Stores conversations as structured facts, beliefs, and summaries across three memory layers with automatic decay and archiving. Eliminates bloated memory files, uncontrolled token growth, and the inability to separate confirmed facts from guesses. Runs a 7-stage consolidation cycle with keyword and entity indexing so agents recall the right context fast.
npx skills add ktao732084-arch/openclaw_memory_supersystem-v1.0 --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install Memory System?โ–ผ

Run `npx skills add ktao732084-arch/openclaw_memory_supersystem-v1.0 --all -g -y` in your terminal to install it globally, then run `python3 src/memory.py init` to create the memory directories.

How to give an AI agent long-term memory?โ–ผ

This system captures conversations into structured facts, beliefs, and summaries, then consolidates them daily into a compact snapshot your agent loads each session.

Does it work without an API key?โ–ผ

Yes. All core features run on pure rules with zero token cost; an optional OpenAI-compatible API key only enhances complex content understanding.

How does it prevent memory bloat?โ–ผ

Memories decay over time based on type and importance, low-score entries are archived automatically, and the working memory snapshot is capped at 2000 tokens.

Does it work with OpenClaw?โ–ผ

Yes. It integrates via system prompt file injection, scheduled cron consolidation, and the standard SKILL.md format used by OpenClaw and compatible agents.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’