semantic-memory

Search project memory using vector embeddings and cosine similarity.

47|10|Updated Feb 22, 2026

Search Tags:#knowledge management #project memory #semantic search #information retrieval #ai #vector embeddings

Ready to run

npx skills add https://github.com/gonzalezpazmonica/pm-workspace --skill semantic-memory-gonzalezpazmonica

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: semantic-memory
Source: https://github.com/gonzalezpazmonica/pm-workspace/tree/main/.claude/skills/semantic-memory
Command: npx skills add https://github.com/gonzalezpazmonica/pm-workspace --skill semantic-memory-gonzalezpazmonica

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a powerful semantic search capability over your project's historical data, allowing you to quickly find relevant information based on meaning rather than just keywords.

Core Features & Use Cases

Semantic Search: Query your project's memory using natural language to find related concepts, decisions, and lessons learned.
Knowledge Retrieval: Access insights from agent notes, past decisions, lessons learned, and postmortems.
Use Case: Ask "What did we decide about the authentication module's caching strategy?" and get precise answers drawn from architectural decisions and agent observations.

Quick Start

Ask the semantic memory skill to find information about past decisions regarding the caching strategy.

Frequently Asked Questions about semantic-memory

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

FAQPage Schema

How does semantic search over project memory work?▼

Semantic search over project memory uses vector embeddings and cosine similarity to match natural language queries with stored architectural decisions and notes, retrieving relevant context based on meaning rather than exact keywords.

How do I retrieve context from past architectural decisions and postmortems?▼

To retrieve context from architectural decisions and postmortems, query the project memory using natural language. The system applies vector embeddings to find and return related concepts, lessons learned, and past observations.

Do I need an embedding model to perform semantic search on agent notes?▼

Yes, you need a local or API-based embedding model to generate vector embeddings. The system calculates cosine similarity against these embeddings, storing the resulting index data in a lightweight JSON format for retrieval.

What is the best way to search lessons learned without using exact keywords?▼

The best way to search lessons learned without exact keywords is using semantic search. It encodes your natural language query into a vector embedding and retrieves matching project notes based on conceptual similarity.

Where does semantic memory store vector index data for information retrieval?▼

Semantic memory stores vector index data in a lightweight JSON format. This approach allows the information retrieval process to access agent notes and project history without requiring a heavy external database.