cass

Mine past agent sessions to surface repeated prompts and decisions.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/Dunc4nJ/agent-skills --skill cass-dunc4nj
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cass
Source: https://github.com/Dunc4nJ/agent-skills/tree/main/skills/cass
Command: npx skills add https://github.com/Dunc4nJ/agent-skills --skill cass-dunc4nj

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Mine past agent sessions to surface working prompts, decisions, and patterns. Use this to understand which prompts repeatedly led to successful outcomes.

Core Features & Use Cases

  • Surface repeated prompts across multiple agents and projects to reveal proven workflows.
  • Detect ritual prompts by aggregating prompts that appear 10+ times.
  • Provide a discovery workflow that indexes, filters, and links prompts to their source sessions for quick context.

Quick Start

Run the prompt miner on your workspace to surface repeat prompts and their contexts in seconds.

Frequently Asked Questions about cass

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

FAQPage Schema
How do I discover repeatable prompts from past agent sessions?

You can mine past agent sessions to discover repeatable prompts by indexing local session logs. This surfaces working prompts, decisions, and patterns alongside their source contexts for rapid analysis.

What are ritual prompts and how are they detected in agent logs?

Ritual prompts are cross-workspace prompts that repeatedly lead to successful outcomes. They are detected by aggregating your prompt history and flagging prompts that appear 10 or more times across multiple agents and projects.

Do I need external dependencies or cloud tools to analyze local agent logs?

No external dependencies are required. You only need local session logs and standard tooling to run the discovery workflow, which filters and links prompts to their source sessions for quick context.

How do I surface proven workflows across multiple agent projects?

You can surface proven workflows by running a prompt miner on your workspace. It targets cross-workspace prompts and outputs structured results with agent, date, and source_path aggregations to reveal repeatable workflows.

What is the best way to extract context-rich decisions from historical agent prompts?

The best way to extract context-rich decisions is to index local session logs and filter for cross-workspace patterns. This links surfaced prompts directly to their source sessions, providing immediate context for analysis.