Memory Search

This searches your conversation history across sessions using semantic, keyword, and symbol matching. It's built for the AI Maestro platform and uses CozoDB for vector search. The workflow is straightforward: before starting any task, search memory for relevant context from previous discussions. You get four modes (hybrid, semantic, term, symbol) depending on whether you're looking for concepts, exact names, or code identifiers. The real value is when you pair it with docs-search and graph-query to build complete context. Requires the AI Maestro stack running locally, so there's setup involved, but if you're already in that ecosystem it slots right in.

skills.sh installs: 139
GitHub stars: 33.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add davila7/claude-code-templates --skill memory-search --agent claude-code

Installs into .claude/skills of the current project.

Files

AI Maestro Memory Search

Search your conversation history using semantic, keyword, and symbol matching. Recall past decisions, discussions, and context across sessions. Part of the AI Maestro suite.

Prerequisites

Requires AI Maestro running locally. Memory indexing uses CozoDB for vector search.

# Install memory tools
git clone https://github.com/23blocks-OS/ai-maestro-plugins.git
cd ai-maestro-plugins && ./install-memory-tools.sh

Core Behavior

Before starting any task, search memory for relevant context:

Receive instruction -> Search memory -> Then proceed

Commands

Command	Description
`memory-search.sh "<query>"`	Hybrid search (recommended)
`memory-search.sh "<query>" --mode semantic`	Find conceptually related
`memory-search.sh "<query>" --mode term`	Exact text matching
`memory-search.sh "<query>" --mode symbol`	Code symbol matching
`memory-search.sh "<query>" --role user`	Only user messages
`memory-search.sh "<query>" --role assistant`	Only assistant messages