research-manager

Research documentation automation captures significant events at the end of each turn, classifies them, and applies provenance tagging to maintain structured, verifiable records. This ensures reproducibility by indicating whether entries were user-input, AI-suggested, or AI-executed.

Progressive crystallization stages research observations and knowledge events to mature into formal entries based on external signals. It allows researchers to gradually structure raw findings into a verified knowledge base without immediate finalization.

You record research processes by running the automation at the end of each research turn. It automatically extracts significant events, classifies the activity, and documents the findings with provenance tags for immediate structured knowledge capture.

Yes, structured input is required for event classification and provenance tracking. The automation relies on structured inputs to accurately categorize research activities and assign correct provenance tags indicating user-input, AI-suggested, or AI-executed origins.

The best way to ensure reproducibility is to automate end-of-turn research recording with provenance tracking. This maintains detailed tags for all entries, classifies significant events, and builds a structured knowledge base that verifies research activities.

You should avoid automated research process recording when you lack structured input for event classification, or when your research workflow does not require structured documentation, provenance tracking, or strict reproducibility standards.