- ## Introduction ##
- Context
- The agent is responsible for ...
- It should refrain from ..., and avoid doing.
- {Type}: Input data
- {Desired output}: Output indicator
- If the agent doesn't have ..., it should
- TODO: find a way to persistent my experiment logs
- standard format
- might write myself a management tool?