search-memory

Query and filter project memories by type, project, file, and date.

17|1|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/vasilievyakov/miracle-infrastructure --skill search-memory-vasilievyakov
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: search-memory
Source: https://github.com/vasilievyakov/miracle-infrastructure/tree/main/skills/search-memory
Command: npx skills add https://github.com/vasilievyakov/miracle-infrastructure --skill search-memory-vasilievyakov

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quickly locate relevant entries across observations, decisions, and problems stored in memory, without scanning entire histories manually.

Core Features & Use Cases

  • Filter by type, project, file, and date to surface relevant memories
  • Load only the details for matched entries to save tokens and speed results
  • Browse dossiers for decisions, unresolved problems, and next steps across projects

Quick Start

Run /search-memory to query your project's memory index across observations, decisions, and problems.

Frequently Asked Questions about search-memory

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

FAQPage Schema
How do I search project memory for past decisions and observations?

To search project memory, you can query stored observations, decisions, and problems using a simple syntax. The system applies filters by type, project, file, concept, and date to surface relevant memories without scanning histories manually.

Can I filter project memories by file, date, and concept across multiple projects?

Yes, you can filter project memories across all projects or a single project. The query system supports filtering by entry type, project name, specific file, concept, and date to provide targeted insights.

What is the best way to find unresolved problems stored in project memory?

The best way to find unresolved problems is to browse the project memory dossiers. You can query the memory index specifically for unresolved problems and next steps across your projects to locate the relevant entries.

How does progressive disclosure work when querying project memory?

Progressive disclosure in project memory queries uses a two-tier index approach. It loads only the details for matched entries rather than the entire history, which saves tokens and speeds up search results.

Do I need to scan the entire history to locate a specific decision?

No, you do not need to scan the entire history manually to locate a decision. You can run a memory search query with filters applied by project, file, and date to quickly find the specific decision entry.