memory

Name: memory
Availability: InStock
Author: jayprimer

Store and search project knowledge with semantic embeddings and full-text search.

Updated Jan 4, 2026

Search Tags:#embeddings #memory #sqlite #project-management #semantic-search #knowledge-base #pmc

Ready to run

npx skills add https://github.com/jayprimer/pmc-marketplace --skill memory-jayprimer

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: memory
Source: https://github.com/jayprimer/pmc-marketplace/tree/main/plugins/pmc/skills/memory
Command: npx skills add https://github.com/jayprimer/pmc-marketplace --skill memory-jayprimer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Storing, indexing, and retrieving project knowledge across sessions to preserve institutional memory and accelerate decision-making.

Core Features & Use Cases

Persistent cross-session memory: stores context, decisions, and learnings in a local store.
Hybrid search: supports semantic embeddings and full-text keyword search for flexible retrieval.
Onboarding and knowledge reuse: accelerates new team member ramp-up by recalling project histories and rationale.

Quick Start

Add a memory with content and a category using the pmc memory add command.

Frequently Asked Questions about memory

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

FAQPage Schema

How do I store project knowledge across sessions using SQLite?▼

Store project knowledge across sessions using SQLite by adding context, decisions, and learnings via the 'pmc memory add' command, which persists data locally for later retrieval and team onboarding.

How does semantic search work with project memories?▼

Semantic search with project memories uses OpenAI embeddings to index content, enabling vector retrieval of conceptually related knowledge across sessions without requiring exact keyword matches.

Do I need an OpenAI API key to use semantic search for project memory?▼

You need an OpenAI API key to enable semantic search for project memory, as the system uses OpenAI embeddings when API keys are available, though full-text keyword search remains functional without them.

Can I use full-text keyword search alongside vector search for project knowledge?▼

You can use full-text keyword search alongside vector search for project knowledge via hybrid search, which combines semantic embeddings and keyword matching to retrieve relevant memories across sessions.

What is the best way to retrieve project decisions using category tagging?▼

The best way to retrieve project decisions using category tagging is to assign categories when adding memories, then filter searches by category to efficiently locate specific historical context and rationale.

Are there limitations when searching memories by partial ID?▼

Partial ID lookups for searching memories are supported for efficient retrieval, but exact matches on complete IDs provide the most precise results when managing specific entries in team projects.