To improve the agent's ability to efficiently answer high-level conceptual questions, its information discovery protocol will be updated. The agent's previous process was inefficient, requiring multiple queries and user feedback to identify the most important architectural documents.
Goal
Modify AGENTS.md to make the agent's information-gathering process more robust and intelligent, ensuring it can identify and prioritize foundational "pillar content" on its own.
Changes to AGENTS.md:
Add tree.json Consultation Step: A new instruction will be added requiring the agent to first consult learn/tree.json for high-level conceptual questions. This will provide the agent with a map of the project's intended information architecture.
Enhance Initial Query Strategy: The "Discovery Pattern" will be updated with a more prescriptive initial step, forcing the agent to query for foundational terms like "benefits", "concept", and "architecture" before narrowing its search.
To improve the agent's ability to efficiently answer high-level conceptual questions, its information discovery protocol will be updated. The agent's previous process was inefficient, requiring multiple queries and user feedback to identify the most important architectural documents.
Goal
Modify
AGENTS.mdto make the agent's information-gathering process more robust and intelligent, ensuring it can identify and prioritize foundational "pillar content" on its own.Changes to
AGENTS.md:Add
tree.jsonConsultation Step: A new instruction will be added requiring the agent to first consultlearn/tree.jsonfor high-level conceptual questions. This will provide the agent with a map of the project's intended information architecture.Enhance Initial Query Strategy: The "Discovery Pattern" will be updated with a more prescriptive initial step, forcing the agent to query for foundational terms like "benefits", "concept", and "architecture" before narrowing its search.