home/productivity/prompt-for-agent-evaluation

Prompt For Agent Evaluation

GPTClaudeDeepSeek··425 copies·updated 2026-07-14
prompt-for-agent-evaluation.prompt
Implement the Edit File Diff Preview feature specified in CONTEXT.md. The confirmation card for edit_file should show a colored unified diff with actual file line numbers instead of the current plain-text preview. Follow the implementation sketch in CONTEXT.md exactly.




Implement the Edit File Diff Preview feature. Spec: in edit_file.py:execute(), after reading current_content, compute a unified diff with correct line numbers using difflib.unified_diff, pass diff_lines into request_confirm(). Update the tool_confirm WS event in agent.py/main.py to include diff_lines. In the frontend, parse it in agent.service.ts and render colored lines (red -, green +) in chat.component.html instead of the plain <pre>.

when to use it

Community prompt sourced from the open-source GitHub repo tmijieux/ai-chat (no explicit license). A "Prompt For Agent Evaluation" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

productivitycommunitydeveloper

source

tmijieux/ai-chat · no explicit license