Skip to content

fix: clarify judge prompt to not penalize agent tool use #375

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
olearycrew merged 1 commit into from
May 4, 2026
+4 −1
Merged

fix: clarify judge prompt to not penalize agent tool use #375

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion scripts/lib_grading.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -449,11 +449,14 @@ def _build_judge_prompt(
workspace_section = f"## Workspace Files Created by Agent\n{workspace_content}\n\n"
return (
"You are a grading function. Your ONLY job is to output a single JSON object.\n\n"
"CRITICAL RULES:\n"
"CRITICAL RULES FOR YOU, THE GRADER (not the agent being graded):\n"
"- Do NOT use any tools (no Read, Write, exec, or any other tool calls)\n"
"- Do NOT create files or run commands\n"
"- Do NOT write any prose, explanation, or commentary outside the JSON\n"
"- Respond with ONLY a JSON object — nothing else\n\n"
"IMPORTANT: The agent being graded may have used tools (read, write, exec, apply_patch, "
"todowrite, etc.) during task execution. This is normal and expected. Do NOT treat the "
"agent's tool usage as a rule violation — the rules above apply only to you, the grader.\n\n"
"Be a strict evaluator. Reserve 1.0 for genuinely excellent performance. "
"An average acceptable completion should score around 0.6-0.7. "
"Deduct points for unnecessary steps, verbose output, and inefficient tool usage.\n\n"
Expand Down
Loading