Claude Code use Cases & Workflows - **Task Decomposition & Delegation** The main agent can delegate parts of a broader coding task to subagents specialized in linting, refactoring, testing, or onboarding new modules. - **Code Review & Refactoring Pipelines** Use subagents in a chain: e.g. “analyzer” -> “optimizer” -> “tester” -> “scribe” to systematically improve a codebase. - **Hybrid Agent Interactions** You can wrap Claude Code within a larger agent orchestration stack (e.g. CrewAI or n8n), invoking subagents for specific tasks. - **Tool & Data Integration** Via MCP, subagents can call external APIs, query databases, fetch context, or run custom logic—bridging Claude with your own backends. - **Iterative Development** Claude Code supports cycles of generation → execution → feedback → revision, fully automated when designed carefully.
# See
- Common workflows - docs.anthropic.com 