Standardizes research data storage and directory conventions to optimize caching and retrieval for AI agents.
The Research Storage skill provides a specialized framework for dokhak agents to manage research data within the Claude Code environment. It implements a strict directory hierarchy and naming convention for chapter and section-based research, ensuring that data collected by researcher agents is easily accessible to writers and designers. By utilizing normalization functions and a sophisticated multi-tier resolution algorithm, the skill prevents redundant web searches and handles legacy naming inconsistencies, creating a reliable cache for high-quality content generation.
主な機能
01Canonical directory structure for organized research caching
02Cross-agent compatibility for research-collector, researcher, and writer roles
03Automated normalization for chapters, sections, and kebab-case slugs
04Multi-tier path resolution to handle naming variations and legacy data
055 GitHub stars
06Standardized Markdown templates for summaries, sources, and research logs
ユースケース
01Enforcing consistent documentation structures across large-scale technical writing projects
02Coordinating data sharing between multiple specialized AI agents in a research pipeline
03Building a persistent research cache to reduce redundant web searches and API costs
