Convert agent list to an enum

[farrelmahaztra]

Note

Replace string-based agent identifiers with a new AgentType enum and update CLI, evaluation flow, examples, and tests to use it consistently.

• Types:
  • Introduce AgentType enum (claude, openai, vllm, litellm, integration_test) in hud/types.py and export it.
• CLI:
  • hud/cli/__init__.py: Use AgentType for agent selection/validation; update interactive choices to return enum values; normalize and cast to AgentType before invoking eval_command.
  • hud/cli/eval.py: Switch eval_command and helpers to accept AgentType; update API key checks and branching logic to compare against enum values; adjust Typer option types.
• Evaluation helpers:
  • Update build_agent, run_single_task, and run_full_dataset to accept and branch on AgentType instead of string literals; propagate enum through config building.
• Examples:
  • examples/run_evaluation.py: Replace Literal["claude","openai"] with Literal[AgentType.CLAUDE, AgentType.OPENAI] and update comparisons accordingly.
• Tests:
  • Update tests in hud/cli/tests/test_eval.py to import and pass AgentType values instead of strings.

^{Written by Cursor Bugbot for commit 0dbb920. This will update automatically on new commits. Configure here.}

[promptless]

📝 Documentation updates detected!

New suggestion: Add AgentType enum API reference for PR #164
Updated existing suggestion: Add comprehensive LiteLLM agent documentation for PR #137 multi-provider support

[dylanbowman314]

@jdchawla29 Do we need to expand AgentType to other parts of the sdk?

[dylanbowman314]

Hmm, do we need this?

[farrelmahaztra]

Ideally no, but hud_console.select doesn't have a generic return type yet based on the values of the choices so by default agent here is str | Literal[AgentType.VLLM]