For Agent_module, only provide format features not enough, need consider as Agent Data store:

The Agent Data store of Carbondata project need have the below features :

1. Use Agents to generate many files, the Agent Data store of Carbondata can manage Agent context(sessions) and relevant output files.
2. When i want to regenerate / rework the existing files, need to retrieve the agent context/session , i can continue to regenerate files(same topics) based on existing high quality historic agent context for optimize the output.

No files, tests, or entry points are named. Start by reviewing the Agent_module and existing CarbonData storage design, then clarify how agent sessions and generated files should be persisted and retrieved. Done means the design and implementation support resuming a session and regenerating related files from its prior context.

Written by the indexing model from the issue text.