Claude Code Configuration & Workflows

Configuring Claude Code for a team: CLAUDE.md memory, custom slash commands, hooks, permission scoping, session management, and wiring it into CI/CD.

1. 3.1CLAUDE.md Hierarchy and OrganizationCLAUDE.md files load from user, project and directory levels; only the project level is shared through version control. Keep them modular with @import and .claude/rules/, and check what loaded with /memory.
2. 3.2Custom Slash Commands and SkillsProject commands and skills are shared through the repository, personal ones live in your home directory. Skill frontmatter controls isolation (context fork), tool access and argument hints; CLAUDE.md is for standards that always apply.
3. 3.3Path-Specific RulesRule files in .claude/rules/ can carry a paths glob in their frontmatter so they load only when Claude works on matching files, which suits conventions that follow file types across the whole tree.
4. 3.4Plan Mode versus Direct ExecutionPlan mode lets Claude explore and design before changing anything; use it for large, multi-file or architectural work and go straight to execution for small, clear changes. Use the Explore subagent to keep noisy discovery out of the main context.
5. 3.5Iterative Refinement TechniquesWhen prose instructions are read inconsistently, show concrete input and output examples, drive the work with tests, let Claude interview you before building in unfamiliar territory, and batch interacting issues into one message.
6. 3.6Claude Code in CI/CD PipelinesIn CI, run Claude Code non-interactively with -p, get machine-readable findings with --output-format json and --json-schema, give it project context through CLAUDE.md, and review with a separate instance from the one that wrote the code.