skills.rest
Sambhav242005
Sambhav242005COMMUNITY
Explore all skills by this author →
WebsiteGitHub

memory-systems

Design persistent semantic memory architectures for agents using vector stores, knowledge graphs, and temporal validity.

Updated Aug 7, 2026
Search Tags:#knowledge graph#agent memory#vector store#graphiti#mem0#memory consolidation#temporal reasoning
One-click install
npx skills add https://github.com/Sambhav242005/Major-Project --skill memory-systems-sambhav242005
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-systems
Source: https://github.com/Sambhav242005/Major-Project/tree/main/.agents/skills/memory-systems
Command: npx skills add https://github.com/Sambhav242005/Major-Project --skill memory-systems-sambhav242005

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires numpy, and includes scripts (resource) and references (resource) components.

What problem does it solve? Agents lose all knowledge when sessions end, forcing users to repeat context and preventing reasoning over accumulated facts. This Skill provides design guidance and working implementations for persistent memory layers that retain entities, relationships, and time-scoped facts across sessions. ## Core Features & Use Cases - Framework Selection Guidance: Compare Mem0, Zep/Graphiti, Letta, Cognee, and LangMem against benchmarks like LoCoMo, LongMemEval, and DMR to pick the right memory architecture. - Layered Memory Design: Choose between working, short-term, long-term, entity, and temporal knowledge graph layers based on retrieval needs rather than defaulting to maximum complexity. - Runnable Reference Implementations: Use the included Python classes (VectorStore, PropertyGraph, TemporalKnowledgeGraph, IntegratedMemorySystem) to prototype memory systems before committing to production frameworks. - Use Case: An agent must remember that a user moved cities in September and answer "where did the user live in March?" correctly. Use the temporal knowledge graph patterns to store facts with validity intervals and run point-in-time queries. ## Quick Start Ask the agent to design a memory architecture for a chatbot that must retain user preferences and entity relationships across sessions, then prototype it with the included memory_store.py script.

Frequently Asked Questions about memory-systems

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

FAQPage Schema
How do I add persistent memory to an AI agent across sessions?▼

Start with the simplest layer that meets your retrieval needs: file-system storage for prototypes, a vector store like Mem0 for semantic search, or a temporal knowledge graph like Zep/Graphiti for facts that change over time. Escalate complexity only when retrieval quality degrades.

Mem0 vs Zep/Graphiti vs Cognee: which memory framework should I choose?▼

Choose Mem0 for fast time-to-production with managed infrastructure, Zep/Graphiti for bi-temporal modeling and relationship traversal, and Cognee for dense multi-layer semantic graphs with customizable ECL pipelines. Letta fits agents needing deep self-introspection over their own memory.

How do I handle facts that change over time in agent memory?▼

Use a temporal knowledge graph that stores valid_from and valid_until timestamps on relationships. Query with point-in-time filters so the agent retrieves the fact valid at the requested date rather than the latest or a stale value.

What benchmarks evaluate agent memory systems?▼

LoCoMo evaluates long-conversation memory retention, LongMemEval measures long-term recall, and DMR scores retrieval accuracy. Treat results as dated, source-specific signals and re-benchmark your own workload before and after architecture changes.

Why does my agent retrieve stale or conflicting memories?▼

Stale memory poisoning happens when outdated facts lack expiry or validity tracking. Implement consolidation that invalidates rather than deletes old facts, prefer the most recent valid_from on conflicts, and surface contradictions to the user when confidence is low.

When should I not use a knowledge graph for agent memory?▼

Avoid graph complexity during prototyping or when queries are simple factual lookups, since file-system or vector storage can match specialized tools on some benchmarks. Add graph structure only when you need multi-hop reasoning, relationship traversal, or time-travel queries.

Related Skills

Explore More Software Engineering Skills→
ehfazrezwanehfazrezwan
5

save-session

Extract and persist session facts and decisions to structured memory.

Community
Intermediate
rustamsrustams

context-fundamentals

Explain context engineering concepts for AI agent systems.

Community
Intermediate
nshaikhsnshaikhs

memory-systems

Identify memory architectures and persistence strategies for cross-session AI agent continuity.

Community
Advanced
khrorekhrore

memory-systems

Implement persistent, queryable agent memory with hybrid retrieval and temporal knowledge graphs.

Community
Advanced