For monorepo projects containing multiple submodules, AGENTS.md uses a hierarchical configuration strategy:
- Global Documentation: Place the base AGENTS.md in the repository root directory to define the generic rules
- Module Proprietary Files: Create targeted command files in each subproject directory
Intelligent bodies will follow the nearest principle when executing:
- pretreatment
/packages/frontendAGENTS.md in this directory is automatically read when the - Backtracking up to parent directory if not found
- User explicit commands always have the highest priority
This design maintains overall consistency while accommodating the specific needs of different technology stack modules, making it an effective solution for managing AI collaboration on large projects.
This answer comes from the articleAGENTS.md: an open-format file providing guidance for AI-programmed intelligencesThe
May not be reproduced without permission:AI productivity tools " How to use AGENTS.md effectively in monorepo project?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil