cass-memory

Extract learned patterns, rules, and anti-patterns from past sessions.

Updated Dec 12, 2025
One-click install
npx skills add https://github.com/eaasxt/Farmhand --skill cass-memory-eaasxt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cass-memory
Source: https://github.com/eaasxt/Farmhand/tree/main/config/skills/cass-memory
Command: npx skills add https://github.com/eaasxt/Farmhand --skill cass-memory-eaasxt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cross-agent learning system. Extracts patterns, rules, and anti-patterns from past sessions to inform current decisions and reduce repeated effort.

Core Features & Use Cases

  • Context-aware retrieval of learned rules and historical context before task execution
  • Aggregation of patterns and anti-patterns from prior sessions
  • Health checks and governance to ensure memory integrity

Quick Start

Run cm context "describe task" to fetch relevant context, rules, and past insights before starting.

Frequently Asked Questions about cass-memory

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

FAQPage Schema
How do I retrieve historical context and learned rules before starting a task?

Retrieve historical context and learned rules by querying a task description through the retrieval API. This surfaces relevant patterns, past insights, and anti-patterns from indexed sessions to inform current work and avoid repeating mistakes.

What is cross-agent memory and how does it improve task planning?

Cross-agent memory aggregates patterns and anti-patterns from past sessions into a shared knowledge base. It improves task planning by providing context-aware retrieval of historical insights, enabling informed decisions across different sessions without repeating mistakes.

Can I extract anti-patterns from past sessions to prevent repeating mistakes?

Yes, the system extracts anti-patterns from past sessions through pattern aggregation. It identifies and surfaces historical mistakes, allowing agents to avoid repeating past errors and ensure memory integrity through health checks and governance.

Does cross-agent learning require session indexing to fetch context?

Yes, cross-agent learning requires session indexing to function. The system indexes past sessions to extract patterns and rules, enabling the retrieval API to fetch relevant context and historical insights before task execution begins.

What are the limitations of using historical patterns for current task guidance?

Historical patterns only reflect past sessions and may not cover novel scenarios. Memory integrity depends on regular health checks and governance to ensure indexed patterns remain relevant and accurate for current cross-agent decisions.