deep-agents-memory

Routes memory and filesystem operations through pluggable StateBackend, StoreBackend, and CompositeBackend backends for Deep Agents.

1|Updated Aug 23, 2025
One-click install
npx skills add https://github.com/laceto/financialtools --skill deep-agents-memory-laceto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-agents-memory
Source: https://github.com/laceto/financialtools/tree/main/.claude/skills/deep-agents-memory
Command: npx skills add https://github.com/laceto/financialtools --skill deep-agents-memory-laceto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Agents often need memory, persistence, and filesystem access; this skill provides pluggable backends to handle ephemeral versus persistent storage and routing.

Core Features & Use Cases

  • StateBackend for short-term memory within a thread
  • StoreBackend for long-term persistence across sessions
  • CompositeBackend for routing memory and file operations across backends
  • FilesystemMiddleware with common file operations (ls, read_file, write_file, edit_file, glob, grep)

Quick Start

Instantiate a Deep Agent with a StateBackend for ephemeral memory and extend it with a StoreBackend or CompositeBackend for persistence and routing as needed.

Frequently Asked Questions about deep-agents-memory

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

FAQPage Schema
How do I add persistent memory to a Deep Agent across sessions?

You add persistent memory to a Deep Agent by configuring a StoreBackend, which handles long-term persistence across sessions. Start with a StateBackend for ephemeral memory and extend it with StoreBackend for persistence.

What is the best way to route file operations to different storage backends?

The best way to route file operations to different storage backends is using a CompositeBackend. It routes memory and filesystem operations across multiple backends, enabling selective data persistence and path routing.

How does filesystem access work with memory backends for Deep Agents?

Filesystem access works with memory backends through FilesystemMiddleware, which handles common file operations like ls, read_file, write_file, edit_file, glob, and grep. These operations route through your configured backends.

Can I use both ephemeral and persistent storage in the same agent?

Yes, you can use both ephemeral and persistent storage in the same agent. Instantiate a Deep Agent with a StateBackend for short-term thread memory and extend it with a StoreBackend or CompositeBackend for long-term persistence.

Do I need a specific runtime to use memory and persistence backends for Deep Agents?

Yes, you need a Deep Agent runtime to use these memory and persistence backends. You also need backend configurations, including optional StoreBackend and routing, to support the Python and TypeScript examples described in the SKILL.md.