memory-retrieval

Retrieve targeted project history from bigmemory hot and cold zones.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/unstun/dqn10 --skill memory-retrieval-unstun
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-retrieval
Source: https://github.com/unstun/dqn10/tree/main/.claude/skills/memory-retrieval
Command: npx skills add https://github.com/unstun/dqn10 --skill memory-retrieval-unstun

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Memory retrieval helps AI fetch relevant project history only when needed, reducing unnecessary context loading and keeping sessions focused.

Core Features & Use Cases

  • On-demand contextual recall from the bigmemory hot and cold zones to support decisions and experiments.
  • Proactive prompts to retrieve decisions, configurations, and milestones from past sessions.
  • Use cases: refreshing memory for ongoing experiments, recalling unresolved decisions, and summarizing past approaches to inform current work.

Quick Start

Instruct memory-retrieval to fetch the most recent unresolved decision from bigmemory using a targeted query.

Frequently Asked Questions about memory-retrieval

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

FAQPage Schema
How do I retrieve past project decisions and configurations during an AI agent session?

You can retrieve past project decisions and configurations by delegating targeted memory retrieval to a memory-retriever agent, which fetches concise project history context from bigmemory hot and cold zones on-demand.

What is on-demand project memory recall and when do I need it?

On-demand project memory recall fetches relevant project history only when needed, reducing unnecessary context loading. You need it for ongoing experiments, unresolved decisions, and summarizing past approaches to inform current work.

Does memory retrieval work with both hot and cold storage zones for project history?

Yes, memory retrieval works with both hot and cold zones in bigmemory. It applies agent-based access to fetch read-only data on ongoing experiments, decisions, configurations, and milestones from past sessions.

What's the best way to fetch the most recent unresolved decision from project history?

The best way to fetch unresolved decisions is instructing memory-retrieval to query bigmemory using a targeted query. This returns concise, read-only outputs suitable for quick reference during active sessions.

Can I use memory retrieval to refresh context for ongoing experiments without loading all past data?

Yes, you can use memory retrieval to refresh context for ongoing experiments without loading all past data. It enforces read-only data retrieval and returns concise outputs, keeping sessions focused on current tasks.

Are there limitations to using agent-based access for project memory retrieval?

Agent-based memory retrieval is limited to read-only data access and applies only when context is required. It cannot modify project history and is invoked exclusively for fetching concise contextual references.