{"id":"b87d9003-44ac-420f-b375-6a8f4ad11cc1","slug":"freeautomation-tech-agent-memory-kit","kind":"mcp","name":"agent-memory-kit","summary":"Memory patterns, context strategies, and persistence recipes for AI agents. Sliding window, vector recall, hybrid memory, MCP memory servers, and more.","owner":"FreeAutomation-Tech","repoUrl":"https://github.com/FreeAutomation-Tech/agent-memory-kit","license":"MIT","licenseTier":"permissive","trust":"community","stars":1,"match":null,"fewStars":true,"format":"mcp","createdAt":"2026-07-13T09:07:36.000Z","addedAt":"2026-10-21T06:24:54.602Z","logoUrl":"https://avatars.githubusercontent.com/u/149557535?v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["memory-knowledge","databases","search"],"tags":["agent-framework","agent-memory","ai-agents","autonomous-agents","context-engineering","crewai","langchain","langgraph"],"runtime":"binary","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-27T18:33:26.135Z","sample":false,"url":"https://conduits.dev/tools/freeautomation-tech-agent-memory-kit","language":"Python","transports":["stdio"],"install":{},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"# Agent Memory Kit — Memory Patterns for AI Agents\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![Patterns: 5](https://img.shields.io/badge/Patterns-5-blue)](#memory-patterns)\n[![Frameworks: 3](https://img.shields.io/badge/Frameworks-3-green)](#framework-guides)\n[![GitHub Repo](https://img.shields.io/badge/GitHub-FreeAutomation--Tech%2Fagent--memory--kit-181717?logo=github)](https://github.com/FreeAutomation-Tech/agent-memory-kit)\n\n> **Memory patterns, context strategies, and persistence recipes for AI agents.** The #1 unsolved pain point in production AI agents — session persistence and context management.\n\n## Why Agent Memory?\n\nAgent memory is the **hottest unsolved problem** in AI engineering (2026). Three repos in this space hit **220K+ stars** in a single week. Every production agent eventually needs:\n\n- Remembering what happened last turn\n- Surfacing relevant past conversations\n- Managing context windows without losing information\n- Persisting state across sessions\n- Choosing what to forget\n\nThis kit solves that. Patterns, not abstractions. Grab what works.\n\n## What's Inside\n\n### Memory Patterns\n| Pattern | Use Case |\n|---------|----------|\n| [Sliding Window](patterns/sliding-window.md) | Keep last N turns — simple, cheap, works for short sessions |\n| [Summary Compression](patterns/summary-compression.md) | Summarize older context to fit within token limits |\n| [Vector Recall](patterns/vector-recall.md) | Semantic search over past conversations using embeddings |\n| [Hybrid Memory](patterns/hybrid-memory.md) | Short-term window + long-term vector recall combined |\n| [Episodic vs Semantic](patterns/episodic-semantic.md) | Separate specific events from general knowledge |\n\n### Memory Strategies\n| Strategy | Description |\n|----------|-------------|\n| [Session Persistence](strategies/session-persistence.md) | Save/load agent state across restarts |\n| [Cross-Session Recall](strategies/cross-session-recall.md) | Remember user preferences and past sessions |\n| [Context Windowing](strategies/context-windowing.md) | Manage context windows for long-running agents |\n| [Cache-Aware Design](strategies/cache-aware-design.md) | Maximize prompt caching for memory operations |\n\n### MCP Integration\n| Guide | Description |\n|-------|-------------|\n| [MCP Memory Server Config](mcp/memory-server-config.md) | Configure MCP memory servers for agent persistence |\n| [MCP Memory Integration](mcp/mcp-memory-integration.md) | Connect MCP tools to agent memory systems |\n\n### Memory Prompts\n| Prompt | Use Case |\n|--------|----------|\n| [Memory Injection](prompts/memory-injection.md) | System prompt patterns for injecting memory context |\n| [Context Summarization](prompts/context-summarization.md) | Prompts for summarizing conversation history |\n| [Recall Instructions](prompts/recall-instructions.md) | Instructions for agents to recall and use past info |\n\n### Framework Guides\n| Framework | Description |\n|-----------|-------------|\n| [LangGraph Memory](frameworks/langgraph-memory.md) | Memory patterns in LangGraph agents |\n| [LlamaIndex Memory](frameworks/llamaindex-memory.md) | Memory in LlamaIndex workflows |\n| [CrewAI Memory](frameworks/crewai-memory.md) | Memory in CrewAI multi-agent systems |\n\n## Quick Start\n\n```bash\ngit clone https://github.com/FreeAutomation-Tech/agent-memory-kit.git\ncd agent-memory-kit\n\n# Browse memory patterns\nls patterns/\n\n# Use a pattern\ncat patterns/sliding-window.md\n\n# Run an example\npython examples/chat-memory.py\n\n# Test memory patterns\npython -m pytest tests/\n```\n\n## Structure\n\n```\npatterns/            # Core memory patterns\nstrategies/          # Memory management strategies\nmcp/                 # MCP memory server integration\nprompts/             # System prompts for memory\nframeworks/          # Framework-specific guides\nexamples/            # Runnable Python examples\ntools/               # Memory analysis tools\ntests/               # Test files\n```\n\n## Usage\n\n```bash\n# Read a pattern\ncat patterns/hybrid-memory.md\n\n# Try the chat memory example\npython examples/chat-memory.py\n\n# Analyze your context usage\npython tools/context_analyzer.py --log conversation.log\n\n# Test memory patterns\npython tools/memory_test.py --pattern sliding-window\n```\n\n## Related Repositories\n\n- [claude-marketing-kit](https://github.com/FreeAutomation-Tech/claude-marketing-kit) — Claude Marketing Skills + Python engine (copy, CRO, analytics, email, social, brand)\n- [claude-seo-kit](https://github.com/FreeAutomation-Tech/claude-seo-kit) — Claude SEO Skills + Python engine (audits, schema, sitemap, hreflang, GEO)\n- [agent-security-kit](https://github.com/FreeAutomation-Tech/agent-security-kit) — Security patterns for AI agents\n- [claude-prompt-kit](https://github.com/FreeAutomation-Tech/claude-prompt-kit) — Prompt engineering patterns\n\n## License\n\nMIT — see [LICENSE](LICENSE).\n\n---\n\n*If you find this useful, please consider giving it a star ⭐ — it helps others discover it too!*\n\n*Thank you for your support! 🙏*\n"}