agent-memory

Capture and track agent expertise across sessions using a structured knowledge registry.

59|12|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/etylsarin/opencastle --skill agent-memory-etylsarin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-memory
Source: https://github.com/etylsarin/opencastle/tree/main/src/orchestrator/skills/agent-memory
Command: npx skills add https://github.com/etylsarin/opencastle --skill agent-memory-etylsarin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Track which agents have expertise with which files, patterns, and tools across sessions to improve task delegation and context.

Core Features & Use Cases

  • Maintain a living memory of agent strengths, weaknesses, and familiarity to inform task assignment.
  • Build a cross-session knowledge graph linking agents, files, and actions to improve collaboration and traceability.

Quick Start

Enable agent-memory tracking in your project to start recording agent expertise across sessions.

Frequently Asked Questions about agent-memory

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

FAQPage Schema
How do I track agent expertise across multiple sessions?

Track agent expertise across sessions by enabling a structured knowledge registry that records agent strengths, weaknesses, and familiarity with specific files and patterns for future reference.

What is cross-session agent memory and when do I need it?

Cross-session agent memory is a structured knowledge graph linking agents, files, and actions to maintain collaboration context. You need it when delegating tasks to agents with proven domain familiarity.

How do I route tasks to agents based on their past performance?

Route tasks by applying a memory system to delegation decisions, matching incoming tasks to agents with proven strengths and domain familiarity recorded in a structured knowledge registry.

Can I build a knowledge graph linking agents to the files they modify?

Yes, you can build a cross-session knowledge graph that links agents, files, and actions to improve collaboration and traceability across your project's development lifecycle.

What are the limitations of using a knowledge registry for agent delegation?

Limitations include managing registry size through defined pruning rules and relying on safe data handling triggers, which require consistent memory updates to maintain accurate delegation matching.