AI Coding Agent

About

• AI agent designed for software engineering tasks
• Can understand codebases, write and modify code, run tests, and fix bugs
• Uses LLMs together with tools such as terminals, editors, and version control
• Can work autonomously across multiple steps to complete a task
• Commonly evaluated using benchmarks such as SWEbench
• Examples include Claude Code, Codex, Gemini CLI, and GitHub Copilot

See also

• AI Agent | LLM
• SWE-bench