//
Busque em todas as páginas da documentação
8 pages in this section.
Explains the shift from autocomplete to autonomous multi-file coding agents and how that reshapes daily engineering work.
A first look at an agent reading, editing, and testing code in a real repo through a bounded tool loop.
Quick-reference for agents that review pull requests for bugs, style, security issues, and policy - without replacing human judgment.
Explains the read-reproduce-fix-verify loop behind autonomous bug-fixing agents and when to trust it.
Quick-reference for agents that generate, refresh, and answer questions about codebase docs grounded in real structure and tests.
Ten practices for scoping coding-agent use cases so humans still review the risky parts while agents handle verifiable work.