AI agents may need cause-and-effect notes, not just timelines
Giving an a full can still lead to wrong answers. The missing piece is often not more information, but the ability to find and connect the right parts of the information. A model may find that a team moved to OAuth2, but miss the earlier discussion that caused the change.
It may also find the right sequence of events and then invent the reason behind them. A better approach is to store history as events, decisions, and clear cause-and-effect links between them. Then, when asked why the team moved to OAuth2, the can follow the chain from the incident to the decision and rollout.
The links need to be conservative: one thing happening before another does not prove it caused it, so a link should only exist when the record supports it.
Key points
- A long does not guarantee that an will find the right reason.
- Larger do not fully solve the problem of choosing the right context.
- Storing events, decisions, and cause-and-effect links can make later answers more accurate.
- For an OAuth2 , the useful record is the chain from issue to decision to rollout.
- Only create cause-and-effect links when the notes clearly support them.