When should I trigger the Session Intelligence Harvester?

You should use it after a session where you corrected format drift, added new validation checks, identified a significant failure mode, or touched multiple files with similar pattern corrections.

Does this skill automatically edit my files?

Yes, the skill is designed to take action by reading target files, applying precise edits to encode new intelligence, and verifying the changes before committing.

What is session harvesting in Claude Code?

Session harvesting is the process of analyzing a productive AI session to extract learnings, patterns, and corrections, then encoding them into the project's permanent documentation to prevent future errors.

What are RII components?

Reusable Intelligence Infrastructure (RII) refers to files like CLAUDE.md, agent profiles, and command files that store the project's context, protocols, and behavioral rules.

Session Intelligence Harvester

Name: Session Intelligence Harvester
Author: Rayder-23

byRayder-23

0•

Productivity & Workflow

Encodes session-specific learnings and corrections into permanent organizational knowledge to prevent recurring errors and pattern drift.

The Session Intelligence Harvester is a specialized skill designed to automate the capture of insights from productive development sessions. It systematically identifies corrections, format drifts, and newly discovered patterns, routing them to the appropriate Reusable Intelligence Infrastructure (RII) components such as CLAUDE.md or agent-specific guidelines. By transforming one-time fixes into permanent architectural updates, it ensures that your AI environment improves with every session, creating a self-reinforcing loop of organizational intelligence that prevents failure modes from repeating.

Key Features

01Automatic Project History Record (PHR) generation and structured commits

02Intelligent routing of learnings to RII components like CLAUDE.md and Agent files

03Validation against canonical sources and project format specifications

04Automated session analysis for error patterns and architectural drift

050 GitHub stars

06Direct multi-file implementation of updates with precise placement

Use Cases

01Updating agent behavioral protocols based on newly discovered failure modes

02Encoding new coding standards after fixing recurring format errors

03Standardizing project workflows by converting session-specific fixes into global rules

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rayder-23/hackathon-1_physical-ai-book session-intelligence-harvester

For use in Claude.ai and ChatGPT

Download Skill