Is the chat agent/model chosen in Copilot Chat important?

[emma284]

Hi community, I have a doubt about the way code is generated with Spec Kit.

When I initialized the project, I did it with
specify init --here --ai copilot

Then, I ran the commands (/constitution, /specify, /plan, etc.) in Copilot Chat choosing Claude Sonnet 4.5 as AI Agent and Spec Kit started to work.
But, what would happen if I choose Grok Code Fast? The output (code) could be different, or the selected model is not important on how Spec Kit works?

I only found this reference to agents in .md files

3. **Agent context update**:
   - Run `.specify/scripts/bash/update-agent-context.sh copilot`
   - These scripts detect which AI agent is in use

so i'm assuming only matter agent selected on initialize.

Thanks in advance.

[itinsecurity]

No, the Spec-kit setting, AFAIK relates to the agent (or "frontend", if you will). Basically so it knows how to relate to your user interface. Which model Copilot calls to do the actual work doesn't matter to spec-kit (it just matters for the cost and quality of your output)