mem-search

Search persistent cross-session memory records to recover past work and decisions.

Updated Apr 2, 2026
One-click install
npx skills add https://github.com/gyalamanch001a/pur-new --skill mem-search-gyalamanch001a
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mem-search
Source: https://github.com/gyalamanch001a/pur-new/tree/main/.agents/skills/mem-search
Command: npx skills add https://github.com/gyalamanch001a/pur-new --skill mem-search-gyalamanch001a

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you quickly recover what was done in previous sessions so you can answer questions like what changed, how an issue was solved, or where a decision came from.

Core Features & Use Cases

  • Cross-session search: Look up past observations, sessions, and prompts by topic, date, project, or type.
  • Context expansion: Open a timeline around a useful result to understand what happened before and after it.
  • Targeted retrieval: Fetch only the relevant observation records instead of loading everything at once.
  • Use case: A developer asks whether a bug was already fixed last week and uses the search, timeline, and fetch workflow to find the exact prior solution.

Quick Start

Ask the skill to search your previous sessions for the topic you care about, then review the returned IDs and fetch the most relevant records for full context.

Frequently Asked Questions about mem-search

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

FAQPage Schema
How do I search previous sessions to find out how an issue was fixed?

Search previous sessions by topic, date, project, or type to recover past fixes. Review the returned IDs, expand the timeline around useful results, and fetch specific observation records for full context.

What is cross-session memory retrieval for project history?

Cross-session memory retrieval searches persistent records to recover past work and decisions. It applies indexed search across observations, timelines, and prompts to answer questions about what changed or where a decision came from.

Can I filter knowledge retrieval by project and date constraints?

Yes, you can filter knowledge retrieval using project filters and date or type constraints. This targeted retrieval fetches only relevant observation records instead of loading entire session histories at once.

How do I expand a timeline around a past observation to see context?

Apply timeline expansion to open a chronological view around a useful search result. This context expansion helps you understand exactly what happened before and after a specific past observation or decision.

What is the best way to recover decisions from past project sessions?

The best way to recover decisions is to search persistent cross-session memory records by topic. Use the search, timeline expansion, and batched observation retrieval workflow to locate the exact prior solution or decision context.

Do I need to load all session history at once to find a past fix?

No, you do not need to load all session history at once. The system uses targeted retrieval with batched observation fetching and project filters to return only the relevant records for your specific query.