MCPs you wire into your main session do not cross into spawned sub-agents. The agent reports Linear unavailable and invents its scope from prompt text.
Context
What to feed the AI so it stays productive — codebase grounding, project conventions (AGENTS.md), persistent decisions, context-window limits, session resets.
You described your project in English. The agent produced generic code that matches none of it. Stop describing. Mention the file. Run the command.
A simple setup guide for Claude Cowork. Connect your files, tools, and preferences so Claude can do real work alongside you.
You typed the rule in chat three times and the agent still broke it. Chat is for the current task. Project rules live in AGENTS.md. Yours is empty.
Three things break an agent session: a wrong turn, a drifted conversation, a thread too big. Three moves fix each: Edit, Restore, Handoff. Picking the wrong one wastes time.
You asked for the same feature twice and got two different implementations. The problem is not the model. Your decisions only live in chat history.
The model did not get worse. Your context filled up. Old noise now fights your new instructions, and the noise is winning.
