session-health

You can diagnose AI session quality by running a self-diagnostic that evaluates context load, scope coherence, and quality degradation signals, recommending whether to continue, summarize, checkpoint, or restart the session.

Scope drift in AI context management happens when active work deviates from the original goal. A scope coherence check detects this drift by comparing the initial objective against current tasks to prevent response quality degradation.

You should checkpoint or summarize an AI coding session when a self-diagnostic detects high context load, scope drift, or quality degradation signals like hedging language and repetitive tool calls, indicating diminishing context sharpness.

You can check for quality degradation signals by monitoring hedging language, repetition, tool call quality, and answer specificity. This evaluates whether the AI's context remains sharp and focused for complex refactoring tasks.

Yes, you can assess team members' learning files and check for uncommitted work during a session health diagnostic if a team configuration exists, ensuring comprehensive context management and safeguarding developer productivity.