Save and retrieve context packages across agent sessions via bazinga_db.py.
npx skills add https://github.com/mehdic/ARTK --skill bazinga-db-context
Please help me install this Agent Skill. Skill: bazinga-db-context Source: https://github.com/mehdic/ARTK/tree/main/.claude/skills/bazinga-db-context Command: npx skills add https://github.com/mehdic/ARTK --skill bazinga-db-context
Context management for AI agents often requires consistent sharing of contextual knowledge, error patterns, and learning strategies across runs and team members. This skill provides a centralized mechanism to save, retrieve, and reason about context packages to improve repeatability and collaboration.
Run the bazinga-db script to save a new context package: python3 .claude/skills/bazinga-db/scripts/bazinga_db.py --quiet save-context-package "<session_id>" "<group_id>" "<package_type>" "<source_file>" "<producer_agent>" '<target_agents_json>' "<priority>" "<content>"
High-intent search queries and answers about installing and using this skill.
You can share context and error patterns across AI agent sessions by using a centralized context package system that saves, retrieves, and tracks learning strategies for reuse. This skill manages those packages to improve cross-agent collaboration.
To save a context package, run a Python script passing the session ID, group ID, package type, source file, producer agent, target agents, priority, and content. This stores the context for later retrieval by other agents.
Yes, Python 3 is required to execute the context package management script. The script handles saving and retrieving context, error patterns, and learning strategies for cross-agent knowledge sharing.
The best way to track context consumption is to use a centralized context management system that automatically records reference updates and usage across sessions. This ensures consistent knowledge sharing and repeatability.
Yes, you can capture and reuse error patterns by saving them into context packages. Agents retrieve these packages in later sessions to apply previously learned strategies and accelerate iteration.
bhutanoInstall and configure a persistent Claude Code usage bar across platforms.
mehdicAssemble relevant context packages for agent spawns within token budgets.
xiaoquqiTrack ML experiments and log metrics with Weights & Biases.
notqueParse and store user-provided error patterns with solutions in the learning database.
