compound

Embed session learnings into system conventions, skills, and patterns.

46|5|Updated Mar 7, 2026

Search Tags:#skill development #research #learning #convention #system improvement #fleet audit

Ready to run

npx skills add https://github.com/victoriacity/openakari --skill compound-victoriacity

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: compound
Source: https://github.com/victoriacity/openakari/tree/main/.claude/skills/compound
Command: npx skills add https://github.com/victoriacity/openakari --skill compound-victoriacity

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill transforms individual session work into accumulated system advantage by embedding learnings into conventions, skills, and patterns, preventing repetitive mistakes and accelerating progress.

Core Features & Use Cases

Learning Consolidation: Captures non-obvious facts, failure modes, and effective techniques discovered during a session.
Task Generation: Creates new tasks for unresolved questions, failed criteria, or multi-phase gaps identified in experiments.
System Improvement: Updates conventions, skills, and patterns based on practical session experience.
Fleet Audit: Reviews recent fleet worker sessions for quality and identifies areas for improvement.

Quick Start

Run the compound skill to embed session learnings into system conventions and patterns.

Frequently Asked Questions about compound

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

FAQPage Schema

How do I embed session learnings into system conventions and skills?▼

To embed session learnings into system conventions, you can run a dedicated skill that analyzes your session work for non-obvious facts, failure modes, and useful techniques, then automatically updates your patterns to create accumulated advantage.

What is the best way to analyze session work for failure modes and convention friction?▼

Analyzing session work for failure modes and convention friction involves reviewing your recent work to capture effective techniques and identify areas where existing patterns caused resistance, allowing you to generate tasks for unresolved questions.

Do I need Git and Bash to audit fleet worker sessions for quality?▼

Yes, you need Git and Bash to audit fleet worker sessions, as Git is required for diffing changes and Bash is necessary for analyzing logs to detect gravity candidates and identify quality improvement areas.

How do I generate tasks for unresolved questions found during development research?▼

You can generate tasks for unresolved questions by analyzing your session experiments to identify multi-phase gaps, failed criteria, and research questions, then embedding these findings as actionable tasks within your system.

Can I update system patterns based on practical session experience automatically?▼

You can update system patterns based on practical session experience by analyzing the work completed and embedding the discovered effective techniques directly into your skills and conventions to prevent repetitive mistakes.