session-search

Retrieve past decisions and sessions from an indexed SQLite FTS5 memory store.

1|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/Largo2z9/phantomos --skill session-search-largo2z9
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-search
Source: https://github.com/Largo2z9/phantomos/tree/main/.skills/skills/session-search
Command: npx skills add https://github.com/Largo2z9/phantomos --skill session-search-largo2z9

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the challenge of locating past conversations, decisions, learnings, and historical context across a workspace without manually searching scattered documents.

Core Features & Use Cases

  • Narrative Memory Search: Searches indexed session history, decisions, snapshots, learnings, and events to retrieve relevant past context.
  • Ranked Historical Retrieval: Returns readable results with source, date, brand, and highlighted snippets instead of raw database output.
  • Use Case: When an operator asks what was decided about a campaign strategy months ago, use this Skill to find the relevant sessions and summarize the available historical evidence.

Quick Start

Ask the session-search skill to find what was previously discussed or decided about a specific topic in the workspace history.

Frequently Asked Questions about session-search

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

FAQPage Schema
How do I search past workspace decisions and session history?

You can search past workspace decisions by querying an indexed SQLite FTS5 narrative memory store using specific terms to retrieve historical context, session records, and learnings. The system returns ranked results with source, date, brand, and highlighted snippets.

What is narrative memory search for historical context discovery?

Narrative memory search is the process of retrieving indexed session history, decisions, snapshots, learnings, and events from a workspace. It applies to historical context discovery tasks involving brand operations, previous discussions, and session records.

Can I filter historical retrieval by brand and date in SQLite FTS5?

Yes, historical retrieval supports filtered querying by specific terms, type, brand, and date within the SQLite FTS5 memory store. This allows targeted discovery of past decisions and events relevant to specific operational contexts.

Does session search require an indexed SQLite FTS5 memory store?

Yes, retrieving past workspace decisions and session records requires access to an indexed SQLite FTS5 memory store. This database infrastructure is necessary to perform the narrative memory search and return ranked historical results.

What is the best way to find what was previously decided about a campaign strategy?

The best way to find what was previously decided about a campaign strategy is to use a session search tool to query the workspace memory index. It summarizes available historical evidence from past sessions and decisions.

What format do historical retrieval results return when searching workspace memory?

Historical retrieval results return readable outputs containing source, date, brand, and highlighted snippets instead of raw database output. This format ensures past decisions and learnings are immediately usable for operators.