Task
Test generation
OpenAI Claude Gemini
Coding Agent
Codebase
AGENTS.md
src
agents
tools
bash.py
edit.py
Coding Tools
Shell
Edit
Grep
Web
Code Changes
Generated Tests
+287 -22
tests
agents
conftest.py
test_bash.py
test_edit.py
Evaluate
Quality Checks
G/W/T Pass Tests
test_lists_repo_and_sees_docs
test_reads_file_and_captures_stderr
test_directory_view_shows_tree
test_edits_and_undoes_working_copy