BeClaude

agentware

New
12GitHub TrendingGeneralby r5rana

Framework that makes your LLM self-aware, loops with self-learning and persistent memory that's private. Open-source & free.

First seen 6/27/2026

Summary

Agentware is an open-source framework that equips LLMs with self-awareness, persistent memory, and self-learning capabilities, enabling them to autonomously improve over time.

  • It provides a private, local memory system that allows agents to retain context across sessions, making it ideal for building long-running, adaptive AI assistants without relying on external cloud services.

Install & Usage

1
Create the agents directory
mkdir -p .claude/agents
2
Save the agent file

Add the configuration to .claude/agents/agentware.md

3
Invoke with @agent-name
@agentware

Use Cases

Build a personal AI assistant that remembers your preferences and past conversations across sessions.
Create a self-improving coding agent that learns from its own mistakes and adapts its behavior over time.
Develop a research assistant that accumulates knowledge from multiple queries and refines its answers based on previous interactions.
Implement a customer support bot that retains user history and continuously improves its responses without manual retraining.
Design a learning companion that tracks a student's progress and adjusts its teaching style based on past performance.
Construct a private, offline-capable agent for sensitive data processing where no external memory services are allowed.

Usage Examples

1

/agentware init --memory persistent --name my-assistant

2

/agentware learn from my previous conversations and improve your responses

3

Using agentware, create a self-learning agent that remembers my coding preferences and suggests improvements.

View source on GitHub
agent

Security Audits

LicenseUnknownSourceWarnRepositoryPass

Frequently Asked Questions

What is agentware?

Agentware is an open-source framework that equips LLMs with self-awareness, persistent memory, and self-learning capabilities, enabling them to autonomously improve over time. It provides a private, local memory system that allows agents to retain context across sessions, making it ideal for building long-running, adaptive AI assistants without relying on external cloud services.

How to install agentware?

To install agentware: create the agents directory (mkdir -p .claude/agents), then add the config to .claude/agents/agentware.md. Finally, @agentware in Claude Code.

What is agentware best for?

agentware is a agent categorized under General. It is designed for: agent. Created by r5rana.

What can I use agentware for?

agentware is useful for: Build a personal AI assistant that remembers your preferences and past conversations across sessions.; Create a self-improving coding agent that learns from its own mistakes and adapts its behavior over time.; Develop a research assistant that accumulates knowledge from multiple queries and refines its answers based on previous interactions.; Implement a customer support bot that retains user history and continuously improves its responses without manual retraining.; Design a learning companion that tracks a student's progress and adjusts its teaching style based on past performance.; Construct a private, offline-capable agent for sensitive data processing where no external memory services are allowed..