hmem-read

Load and query long-term memory across projects via hmem APIs.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/YuryZZZ/opencode-global --skill hmem-read-yuryzzz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hmem-read
Source: https://github.com/YuryZZZ/opencode-global/tree/main/skills/hmem-read
Command: npx skills add https://github.com/YuryZZZ/opencode-global --skill hmem-read-yuryzzz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Load and manage long-term memory to preserve context across sessions, enabling continuity when switching projects or resuming prior work.

Core Features & Use Cases

  • Cross-project memory access with load_project and read_memory to activate on-demand context.
  • Memory health auditing: stale detection, memory_stats, find_related, and related-entry retrieval.
  • Cross-session continuity: resume work after pauses, recall past decisions, and leverage L1/L2 summaries for faster task execution.

Quick Start

Load the active project's memory to begin with the full briefing and recent exchanges.

Frequently Asked Questions about hmem-read

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

FAQPage Schema
How do I maintain context across sessions when switching projects?

You can detect stale memory entries by running proactive memory health audits. The system applies stale detection alongside memory_stats and find_related APIs to evaluate memory freshness and identify outdated cross-project references.

What is the best way to retrieve related references during complex AI workflows?

The best way to retrieve related references is by invoking find_related and read_memory APIs. This rapidly loads relevant briefings, O-entries, and cross-project references on demand to preserve operational continuity.

Can I use cross-project memory access to recall past decisions after a pause?

Yes, cross-project memory access supports recalling past decisions after pauses. By calling load_project, you activate on-demand context and leverage stored L1/L2 summaries for faster task execution.

Does long-term memory management work without external dependencies?

Long-term memory management operates with no external dependencies. It internally invokes read_memory, write_memory, and related hmem APIs to handle cross-project reference retrieval and stale detection natively.

Why does cross-project memory retrieval return stale entries?

Cross-project memory retrieval may return stale entries if proactive memory health auditing is not regularly performed. Running memory_stats and stale detection helps identify and clear outdated references from the active project context.