skills.rest
dotlab-hq
dotlab-hqOFFICIAL
Explore all skills by this author →
GitHub

deep-agents-memory

Routes AI agent memory and filesystem access through pluggable backends and middleware.

Updated Apr 12, 2026
Search Tags:#routing#backend#memory#persistence#filesystem#composite#deep-agents
One-click install
npx skills add https://github.com/dotlab-hq/torque --skill deep-agents-memory-dotlab-hq
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-agents-memory
Source: https://github.com/dotlab-hq/torque/tree/main/.agents/skills/deep-agents-memory
Command: npx skills add https://github.com/dotlab-hq/torque --skill deep-agents-memory-dotlab-hq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Agents often need to store ephemeral data during a task or persist memory across sessions. This skill provides pluggable backends (StateBackend, StoreBackend, CompositeBackend) and a filesystem middleware to handle memory, persistence, and file operations securely.

Core Features & Use Cases

  • StateBackend for short-lived memory within a thread.
  • StoreBackend for long-term persistence across sessions.
  • CompositeBackend to route paths between backends.
  • FilesystemMiddleware for common file operations (ls, read_file, write_file, edit_file, glob, grep).

Quick Start

Configure a Deep Agent with the desired backend (StateBackend, StoreBackend, or CompositeBackend) and perform a sample memory operation such as saving to /memories/style.txt.

Frequently Asked Questions about deep-agents-memory

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

FAQPage Schema
How do I persist AI agent memory across sessions?▼

To persist AI agent memory across sessions, you configure a long-term storage backend like StoreBackend, which retains memory and data beyond the lifespan of a single thread or session.

What is the best way to manage both short-lived and persistent memory for AI agents?▼

The best way to manage mixed memory lifespans is using composite routing. A CompositeBackend routes paths between ephemeral storage for short-lived thread data and persistent backends for cross-session memory.

Can I integrate filesystem operations like reading and writing files into agent memory workflows?▼

Yes, you can integrate filesystem operations into agent memory workflows using filesystem middleware. It handles common file operations such as ls, read_file, write_file, edit_file, glob, and grep securely.

How do I route memory storage to different backends within the same agent workflow?▼

You can route memory storage to different backends by configuring explicit routing rules with a CompositeBackend. This allows your agent workflow to direct paths dynamically between ephemeral and persistent storage.

Does this memory routing approach work for production-grade agent workflows?▼

Yes, this memory routing approach is designed for production-grade agent workflows. It provides modular backend selection and filesystem middleware integration to ensure robust agent memory and data management.

Related Skills

Explore More Software Engineering Skills→
dawsonblockdawsonblock

hermes-agent

Configure and extend Hermes Agent across GUI and code execution backends.

Community
Advanced
githubgithub
4.9k

gh-agent-session

Create and manage GitHub Copilot agent sessions via gh agent-task.

Official
Intermediate
jxtngxjxtngx
2

deep-agents-memory

Route agent memory operations across ephemeral and persistent storage backends.

Community
Advanced
sakamoto-family-smilesakamoto-family-smile

knowledge-ops

Store and organize knowledge across local files, MCP memory, vector stores, and Git repositories.

Community
Advanced