Can I adjust the number of memories retrieved?

Yes, you can use the --k flag followed by a number to increase or decrease the number of results returned by the query.

Does this require a specific database setup?

The skill is designed to work with a PostgreSQL instance configured for the Maestro framework to store and query embeddings.

How does Maestro Memory Recall store my past work?

It utilizes a PostgreSQL database paired with BGE embeddings to index and store 'learnings' from your previous sessions, making them searchable by meaning.

What is the difference between vector-only and text-only search?

Vector-only search uses semantic embeddings for higher precision based on concept, while text-only search performs traditional keyword matching.

Maestro Memory Recall

Name: Maestro Memory Recall
Author: scooter-lacroix

byscooter-lacroix

Learning & Documentation

Retrieves semantic learnings and context from past coding sessions to provide persistent project intelligence.

About

Maestro Memory Recall enables Claude to tap into a persistent semantic memory system, allowing it to remember specific implementation patterns, installation steps, and bug fixes across different sessions. By querying a PostgreSQL database using BGE embeddings, this skill surfaces relevant historical context and previous learnings, ensuring that established project conventions are consistently applied and that previously solved problems do not require re-explanation.

Key Features

Semantic search across historical session data
0 GitHub stars
Seamless integration with PostgreSQL and BGE embedding systems
Customizable result limits to control context injection depth
Configurable search modes including vector-only and text-only
Detailed result output with confidence scores and session IDs

Use Cases

Troubleshooting recurring TypeScript errors based on past successful fixes
Retrieving complex installation steps or configuration patterns for internal tools
Recalling specific hook development patterns used in previous sprints

About

Key Features

Semantic search across historical session data
0 GitHub stars
Seamless integration with PostgreSQL and BGE embedding systems
Customizable result limits to control context injection depth
Configurable search modes including vector-only and text-only
Detailed result output with confidence scores and session IDs

Use Cases

Troubleshooting recurring TypeScript errors based on past successful fixes
Retrieving complex installation steps or configuration patterns for internal tools
Recalling specific hook development patterns used in previous sprints