skills.rest
Cheggin
ChegginCOMMUNITY
Explore all skills by this author →
Website𝕏@reagan_hsuGitHub

tiered-memory

Manage hot, warm, and cold memory tiers on disk with TTLs.

1|1|Updated Apr 13, 2026
Search Tags:#memory#persistence#context-management#agent-memory#context-reset#hot-memory#tiered-memory
One-click install
npx skills add https://github.com/Cheggin/request-for-startups --skill tiered-memory
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tiered-memory
Source: https://github.com/Cheggin/request-for-startups/tree/main/skills/tiered-memory
Command: npx skills add https://github.com/Cheggin/request-for-startups --skill tiered-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Three-tier memory system stores context on disk in .harness/memory/{hot,warm,cold} to prevent context pollution and survive resets.

Core Features & Use Cases

  • Hot tier: always loaded (max 2K tokens) and injected into every agent session; contains current task description, active feature checklist, and blockers; auto-summarizes when cap is reached.
  • Warm tier: seven-day TTL; loaded on demand; stores recent architectural decisions, recent GitHub issues, and common error patterns; entries expire after 7 days.
  • Cold tier: permanent; source of completed feature summaries, historical decisions, and post-mortems; surfaced when needed.
  • Promotion/demotion: moves items between tiers as relevance changes (hot-to-warm/cold; cold-to-hot when reactivated).
  • Persistence: integrates with context-reset-handler; PostStart loads hot memory and provides access to warm and cold retrieval.

Quick Start

Initialize and deploy the tiered-memory system in your harness to enable persistent multi-tier context across resets.

Frequently Asked Questions about tiered-memory

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

FAQPage Schema
How do I persist agent memory across context resets?▼

A three-tier memory system preserves knowledge across context resets by writing to disk storage, preventing context pollution and maintaining continuous agent awareness without losing historical data.

What is the best way to manage hot, warm, and cold memory tiers for autonomous agents?▼

The best way to manage tiered memory is to cap hot storage at 2K tokens for current tasks, set a 7-day TTL for warm tier architectural decisions, and use cold tier for permanent post-mortems with promotion and demotion rules.

How does a tiered memory system prevent context pollution in long running agent sessions?▼

Tiered memory prevents context pollution by capping the hot tier at 2K tokens, auto-summarizing when full, and moving less relevant data to warm or cold storage instead of keeping everything in the active context window.

Does tiered memory require on-disk persistence to survive agent session resets?▼

Yes, tiered memory requires on-disk persistence at a designated directory to survive resets. It integrates with a context-reset-handler to load hot memory on startup and provide explicit access to warm and cold retrieval.

When should I promote or demote items between hot, warm, and cold memory tiers?▼

You should promote or demote items between memory tiers as relevance changes, moving completed hot items to warm or cold storage, and reactivating cold items back to hot when they become relevant to the current task.

Related Skills

Explore More Software Engineering Skills→
TheBushidoCollectiveTheBushidoCollective
187

claude-agent-sdk-context-management

Manage agent memory, context, and conversation state in Claude AI agents.

Official
Advanced
ak-eytherak-eyther

memory-management

Consolidate agent memory across hot, warm, and cold tiers after task completion.

Community
Intermediate
Dqz00116Dqz00116
22

knowledge-base-cache

Create and manage layered knowledge bases with hot, cold, and warm caches.

Community
Advanced
wewetv1987-cellwewetv1987-cell

memory-tiering

Organize multi-tier memory into HOT, WARM, and COLD storage tiers.

Community
Intermediate