/Glossary · AI
What is Claude Code?
Claude Code is Anthropic's agentic coding tool that reads your repository, edits files, runs commands, and proposes changes from natural-language instructions. It works in the terminal, in your IDE, and in CI pipelines.
How Claude Code works
Claude Code runs on your machine or on a CI runner and works against the real repository: it searches files, reads code, runs tests and shell commands, and edits what it needs to complete the task. It asks for permission before sensitive actions, and you can tune those permissions per command or per tool in the project settings.
You use it from the terminal, through extensions for VS Code and JetBrains, and in GitHub Actions or other CI systems to review pull requests or work through issues. The CLAUDE.md file, versioned in the repo, gives it persistent context: build commands, conventions, architecture, and the rules your team wants it to follow.
Claude Code with Amazon Bedrock and MCP
Claude Code can use Claude models through Amazon Bedrock instead of the Anthropic API. You turn it on with the CLAUDE_CODE_USE_BEDROCK variable; it picks up credentials from the standard AWS SDK chain and respects your IAM policies, so usage and audit trails stay in your own account.
Through Model Context Protocol, Claude Code connects to MCP servers that expose external tools: AWS documentation, your issue tracker, a read-only database, or your observability console. The agent then queries real data from your environment instead of guessing, limited to the permissions each server defines.
Best practices for teams
Keep CLAUDE.md short and specific, with the commands that validate a change and the project's constraints. Restrict default permissions, use read-only IAM roles for analysis tasks, and make every change land on a branch and a pull request that goes through your usual human review.
For infrastructure, have Claude Code generate the Terraform plan and post it to the pull request before anything is applied. Tracking cycle time and change failure rate shows whether the tool is actually improving delivery or just shifting work onto reviewers.
/FAQ
Questions about Claude Code
Another question? Write to hello@nubent.com and an engineer will answer.
Can I use Claude Code with my AWS account?
Yes. Claude Code supports Amazon Bedrock as a model provider. You set the CLAUDE_CODE_USE_BEDROCK variable and the region, enable Anthropic models in Bedrock, and use your usual AWS credentials, for example an IAM Identity Center profile.
What is the CLAUDE.md file?
It is a Markdown file that Claude Code reads at the start of a session to get project context. It usually covers build and test commands, code conventions, repository structure, and team rules. It can live at the project, subdirectory, or user level.
Can Claude Code make changes without supervision?
It can run multi-step tasks on its own, but permissions are configurable and by default it asks for approval before sensitive actions. For professional teams, the recommended setup is for every result to arrive as a pull request that goes through human review and the CI pipeline.
How we help with Claude Code
Claude Code for teams
Claude Code adoption for engineering teams: setup, CLAUDE.md, permissions, MCP, CI/CD, code review, credential security and measuring the impact on delivery.
Learn moreDevOps with AI agents
DevOps operations assisted by AI agents: Terraform changes by pull request, alert triage, cost reviews, runbooks and postmortems, always with human approval.
Learn moreMCP servers
Custom MCP servers that connect Claude and other agents to your internal systems and AWS, with authentication, least privilege, audit logging and deployment in your account.
Learn moreAI training
AI workshops for engineering teams: Claude Code, agents and RAG on Amazon Bedrock, MCP servers and AI security, with exercises built on your own codebase.
Learn moreNext step
What should we build next on AWS?
30 minutes with an engineer. Concrete next steps, whether or not you work with us.