Eval GPT
Evaluate and rank agent results by metric or LLM judge for an AgentHub session. Use when the user runs /hub:eval or asks to score, compare, or pick a winner among completed AgentHub agents.
Eval GPT
Evaluate and rank agent results by metric or LLM judge for an AgentHub session. Use when the user runs /hub:eval or asks to score, compare, or pick a winner among completed AgentHub agents.
Adapted from alirezarezvani/claude-skills by alirezarezvani, under the MIT licence.
See ATTRIBUTION.md for the exact source and commit.
How to install
ChatGPT GPT builder · ChatGPT
- In ChatGPT, open Explore GPTs › Create, then the Configure tab.
- Name: Eval
- Description: Evaluate and rank agent results by metric or LLM judge for an AgentHub session. Use when the user runs /hub:eval or asks to score, compare, or pick a winner among completed AgentHub agents.
- Paste instructions.md (2,225 characters) into Instructions.
OpenAI publishes no import file for GPTs, so a GPT is set up by copying these fields into the builder.
From the vendor’s docs: help.openai.com
Versions
- 1.0.0 · 2026-10-29 · 3.1 KB
Imported from alirezarezvani/claude-skills at 0496583603e5.
sha256 803714db926e64a651d194dd005907f1af13bbde3708470c648a0f9f2afdbc26