Does this skill track token consumption?

Yes, it includes specialized commands to record and retrieve consumption data, helping you monitor how agents are utilizing resources and context.

How does the error pattern learning work?

The skill saves error signatures and their successful resolutions. As these patterns are reused and prove helpful, their confidence score increases, helping future agents solve similar issues faster.

Can I prioritize what information an agent receives?

Yes, context packages can be assigned priority levels (critical, high, medium, low), and the retrieval command automatically orders them to ensure the most important information is consumed first.

What is a context package in Bazinga DB?

A context package is a structured container used to share specific knowledge—such as research, investigation results, or decisions—between different agents in a multi-agent workflow.

Bazinga DB Context

Name: Bazinga DB Context
Author: mehdic

bymehdic

•

Database Management

Manages cross-agent knowledge sharing, persistent learning patterns, and context consumption for multi-agent AI teams.

About

Bazinga DB Context is a specialized memory and learning skill designed for multi-agent development workflows within Claude Code. It enables seamless knowledge transfer by allowing agents to save and retrieve context packages—such as research findings, architectural decisions, and handoff notes—across different agent spawns. Additionally, the skill implements a persistent learning layer that tracks error patterns and successful strategies with confidence-based scoring, allowing the AI team to become more efficient by referencing past resolutions and proven methodologies during complex coding tasks.

Key Features

Cross-agent context package sharing for research and decisions
Granular context consumption and token usage tracking
Strategy storage for reusable debugging and architectural patterns
2 GitHub stars
Persistent error pattern tracking with confidence-based learning
Priority-based retrieval system for optimized agent initialization

Use Cases

Building a project-specific knowledge base of resolved build and test errors
Optimizing token usage by tracking and managing context consumption across agent spawns
Handoffs between specialized agents (e.g., researcher to developer)

About

Key Features

Cross-agent context package sharing for research and decisions
Granular context consumption and token usage tracking
Strategy storage for reusable debugging and architectural patterns
2 GitHub stars
Persistent error pattern tracking with confidence-based learning
Priority-based retrieval system for optimized agent initialization

Use Cases

Building a project-specific knowledge base of resolved build and test errors
Optimizing token usage by tracking and managing context consumption across agent spawns
Handoffs between specialized agents (e.g., researcher to developer)