An agent should answer questions from governed tables in plain language. What can it use?
Why B is the answer
Genie Spaces
Genie Spaces answer plain-language questions from governed tables, and an agent can call them for structured data.
In detail
- Genie Spaces or the conversational API can retrieve data.
- They answer plain-language questions over governed tables.
- Tell the agent when to use Genie, and when to use documents.
Use case
An illustrative scenario, not a real incident.
A sales lead asks an agent for last quarter's top regions. The agent uses a Genie Space, which answers from governed tables in plain language.
See it in different ways
flowchart TD A["User"] --> B["Supervisor agent"] B --> C["Knowledge agent"] B --> D["Genie Space: structured data"]
sequenceDiagram
participant U as User
participant A as Agent
participant I as Vector index
participant T as Tool
U->>A: Question
alt General policy
A->>I: Search documents
I-->>A: Passages
else Live or per-record fact
A->>T: Call with a key
T-->>A: Current value
end
A-->>U: AnswersequenceDiagram participant P0 as User participant P1 as Agent participant P2 as Genie Space participant P3 as Governed tables P0->>P1: Plain question P1->>P2: Ask Genie P2->>P3: Query P3-->>P1: Result P1-->>P0: Answer
Key point: Genie answers plain questions from governed tables.
Why the other options are wrong
- A. A model with no tools
- A model with no tools cannot reach the tables.
- C. A Knowledge Assistant over PDF exports of the tables
- PDF exports go stale, and sit outside table governance.
- D. A vector index of the column names
- Column names alone contain none of the data.