Claude Code
Anthropic's agentic coding tool for the terminal, IDEs, a desktop app and the web, built around permissions, subagents and MCP.
- Agentic tool
- Vendor-sourced
What is Claude Code?
Claude Code is Anthropic's agentic coding tool. It runs in a terminal, in VS Code and JetBrains extensions, in a desktop app and on the web, and uses Claude models with tools, permissions and subagents.
- Best forTerminal-first developers
- PricingIncluded with paid Claude plans, or usage-based via API billing.
- Main limitationThe documentation reviewed describes Claude models only, so teams wanting a choice of model vendors may prefer a multi-vendor product.
- AlternativesCursor
Claude Code overview: who it is for and what it does.
Claude Code is a coding tool from Anthropic that developers direct in natural language. Anthropic's documentation covers use from the terminal CLI, IDE extensions for VS Code and JetBrains, a desktop app and Claude Code on the web, along with GitHub Actions and a Code Review feature.
The documentation emphasises control and extensibility. It describes a permission system used together with sandboxing for defence in depth, checkpoints, MCP servers for connecting external tools, hooks, skills, plugins and custom subagents. A CLAUDE.md file provides project instructions that are read at the start of every session.
Access is through paid Claude plans (Pro, Max, Team and Enterprise) or usage-based billing through an API key or a chosen cloud provider, according to third-party summaries of Anthropic's documentation. Disclosure: Anthropic also makes the Claude assistant, and this profile applies the same sourcing rules as every other listing.
Main problem it solves
Lets developers delegate multi-step coding tasks, such as exploring a codebase, refactoring and running tests, to an AI tool that works in their own environment under configurable permissions.
Typical customer segment
Individual developers through enterprises, via Pro, Max, Team and Enterprise plans or usage-based API billing.
Target users
- Developers who work in the terminal or want one tool across terminal, IDE and web
- Teams that want to standardise project instructions, permissions and extensions
- Organisations that want centrally managed settings
Primary use cases
- Exploring and explaining a codebase
- Refactoring across files
- Running tasks in parallel with subagents or Git worktrees
- Automating code review and CI tasks
- Connecting external tools through MCP
What makes it distinct
- Runs on several surfaces: terminal, VS Code, JetBrains, desktop and web.
- A documented permission and sandboxing model, with checkpoints.
- Extensibility through MCP, hooks, skills, plugins and subagents.
Claude Code features explained.
Where it runs
- Terminal CLI
- The core interface, installed as a command-line tool and started inside a project directory.
- IDE extensions
- Extensions for VS Code and JetBrains IDEs.
- Desktop app and web
- A desktop app and Claude Code on the web at claude.ai/code, with documented ways to move tasks between web and terminal.
Control and safety
- Permissions
- Rules that block or allow specific tools and commands, for example denying shell access with a deny rule.
- Sandboxing
- Anthropic's documentation describes permissions and sandboxing used together as defence in depth.
- Checkpoints
- Documented as part of how to stay safe when working with Claude Code.
Context and extensibility
- CLAUDE.md project instructions
- A file that Claude Code reads at the start of every session.
- Subagents
- Custom subagents with their own context, tools and permissions that return a summary to the main conversation.
- MCP servers, hooks, skills and plugins
- Connect external tools and customise behaviour; plugins bundle skills, MCP connectors and subagents.
Automation
- GitHub Actions and Code Review
- The documentation covers a Claude Code GitHub Actions integration and a Code Review feature.
- Parallel sessions
- Documentation covers running parallel Claude Code sessions with Git worktrees.
Claude Code pricing and plans.
Structure only; figures not recorded. Checked 15 Oct 2026 against innFactory AI · Anavem.
Pricing model
Included in paid Claude plans (Pro, Max, Team, Enterprise), or usage-based through an API key or a chosen cloud provider.
Free plan
Not documented
Free trial
Not documented
Claude Code plans and prices (USD)
Claude Pro
Price not recorded
Subscription
Claude Code is listed on the Pro plan.
Claude Max
Price not recorded
Subscription
Offers 5x or 20x more usage than Pro, per a third-party summary of Anthropic's documentation.
Team and Enterprise
Price not recorded
Per seat
Usage draws from a per-seat allowance on rolling five-hour and weekly windows, per a third-party summary of Anthropic's cost documentation.
API or cloud provider billing
Usage-based
Per token
Billed through an API key or the chosen cloud provider.
Plan prices are not recorded because they were not reviewed on Anthropic's own pricing pages for this profile. Check Anthropic's pricing page before budgeting.
Prices change. Confirm current plans and terms on the vendor's site before you decide.
Visit official website (external) ↗See alternativesClaude Code integrations, API and platforms.
Important integrations
- MCP servers
- VS Code and JetBrains
- GitHub Actions
- Git worktrees
- Skills, hooks and plugins
Supported platforms
- Terminal (CLI)
- VS Code
- JetBrains
- Desktop app
- Web
API
Anthropic documents an Agent SDK for building agents on the same runtime. API details for Claude Code itself were not reviewed.
Data import and export
Not reviewed for this profile.
Documentation lists commands whose availability depends on platform, plan and environment; for example, the /desktop command is only available on macOS and Windows.
How Claude Code works as an AI product.
Attributes are listed only where documented. We do not infer autonomy levels, certifications or model capabilities.
Classification: Agentic tool. A human-directed tool that can plan, edit and run actions, with the user reviewing or approving changes. Developer: Anthropic.
| Attribute | Details |
|---|---|
| Supported models | Claude models from Anthropic. Usage can be billed through an API key or a chosen cloud provider. [2] |
| Tools and integrations | MCP servers, hooks, skills, plugins and subagents; GitHub Actions; Code Review. [1] [2] |
| Ability to execute actions | Claude Code can edit and refactor code and work in the terminal environment, subject to a permission system. [1] |
| Execution environment | Terminal CLI, VS Code and JetBrains extensions, a desktop app and the web at claude.ai/code. [1] [2] |
| Human-in-the-loop controls | Permissions and sandboxing are used together for defence in depth, and the documentation covers checkpoints and permission configuration. [1] |
| Permissions and access | Permission rules can block access to restricted resources, for example a deny rule on shell commands. Managed settings can be distributed centrally and take precedence over other configuration levels, per a third-party summary. [1] [2] |
| Memory and context | A CLAUDE.md file is read at the start of every session. Subagents have their own context windows and return a summary. Sessions can be resumed. [1] [2] |
| Deployment options | Paid Claude plans (Pro, Max, Team, Enterprise) or usage-based API or cloud-provider billing, across CLI, IDE, desktop and web. [2] |
| Data handling and privacy | According to a third-party summary of Anthropic's documentation, Anthropic does not train generative models on code or prompts sent to Claude Code under commercial terms (Team, Enterprise, API) unless the customer opts in, while consumer users choose whether data is used. Confirm in Anthropic's terms. [2] [3] |
| Security and compliance | Managed settings through MDM, SSO and role-based permissions under Claude for Enterprise are described in a third-party summary. Compliance certifications were not reviewed for this profile. [2] |
| Pricing and usage limits | Subscription usage is limited. Max offers 5x or 20x more usage than Pro, and Team and Enterprise draw from per-seat allowances on rolling five-hour and weekly windows. [3] |
| Languages and platforms | Not documented in the sources reviewed. |
| Limitations and known constraints | Some features depend on plan and platform, and on Team and Enterprise accounts certain features such as channels must be enabled by an admin. [1] [3] |
Primary tasks
- Exploring and refactoring codebases
- Running commands and tests
- Code review
- Parallel and automated coding tasks
Best-fit workflows
- Developers who work in the terminal and want the same tool in their IDE and the web
- Teams that want to standardise permissions, project instructions and extensions
- Workflows that delegate parallel or repetitive tasks to subagents or CI
Balanced and sourced.
Each point cites a source. Editorial summaries are not customer reviews.
Strengths
- Runs in the terminal, VS Code, JetBrains, a desktop app and the web, so one tool can follow a developer across environments. [1]
- Anthropic documents a permission system used with sandboxing, plus checkpoints, which gives teams concrete oversight controls to evaluate. [1]
- Extensible through MCP servers, hooks, skills, plugins and subagents with their own context windows. [1]
- A CLAUDE.md file lets teams give the tool persistent project instructions that load every session. [1]
Limitations
- The documentation reviewed describes Claude models only, so teams wanting a choice of model vendors may prefer a multi-vendor product. [2]
- Subscription usage is limited and measured in rolling five-hour and weekly windows, which heavy users should plan for. [3]
- No free tier was identified in the sources reviewed, so evaluation requires a paid plan or API usage. [2]
- Pricing was not reviewed on Anthropic's own pricing pages for this profile, so cost comparisons need a check of the live page. [3]
Best for
- Terminal-first developers
- Teams that want documented permission and sandbox controls
- Workflows that use subagents, MCP integrations or CI automation
May be less suited to
- Teams that want to switch freely between model vendors in one product
- Developers who want an editor-centred experience as the primary interface
Alternatives to Claude Code.
Claude Code compared with similar ai coding agents.
Summaries come from each profile. This is a side-by-side aid, not a ranking.
| Product | Interface | Model choice | Free tier | Pricing model | Human oversight |
|---|---|---|---|---|---|
| Claude Code | Terminal, VS Code, JetBrains, desktop app, web | Claude models (Anthropic) | Not documented in sources reviewed | Included in paid Claude plans, or usage-based API billing | Permissions, sandboxing and checkpoints documented |
| Cursor Compare | AI code editor (desktop) with cloud agents | OpenAI, Anthropic, Google and Cursor models | Yes (Hobby) | Monthly tiers; usage at model API rates | Not documented in sources reviewed |
Claude Code questions and answers
What is Claude Code used for?
Claude Code is used to explore, edit and refactor codebases and to run commands and tests from the terminal, IDE extensions, a desktop app or the web, under a configurable permission system.
Is Claude Code an autonomous agent?
We classify it as an agentic tool. It can plan and act, but the documentation describes permissions, sandboxing and checkpoints that keep the developer in control, and it does not describe fully autonomous operation.
Does Claude Code have a free plan?
No free tier was identified in the sources reviewed. It is described as part of paid Claude plans or available with usage-based API billing. Confirm on Anthropic's pricing page.
Which models does Claude Code use?
It uses Claude models from Anthropic. Usage can be billed through an API key or a chosen cloud provider.
How does Claude Code compare with Cursor?
Claude Code is terminal-first and also runs in IDEs, a desktop app and the web, using Claude models. Cursor is an editor with models from several vendors. See the Cursor vs Claude Code comparison.
What integrations does Claude Code support?
Documented integrations and ecosystem items include MCP servers, VS Code and JetBrains, GitHub Actions, Git worktrees, Skills, hooks and plugins. API details: Anthropic documents an Agent SDK for building agents on the same runtime. API details for Claude Code itself were not reviewed.
What are the limitations of Claude Code?
The documentation reviewed describes Claude models only, so teams wanting a choice of model vendors may prefer a multi-vendor product. Subscription usage is limited and measured in rolling five-hour and weekly windows, which heavy users should plan for. No free tier was identified in the sources reviewed, so evaluation requires a paid plan or API usage.
Which platforms does Claude Code run on?
Claude Code is documented as available on: Terminal (CLI), VS Code, JetBrains, Desktop app, Web.
Is Claude Code an AI agent or an AI assistant?
We classify Claude Code as: Agentic tool. A human-directed tool that can plan, edit and run actions, with the user reviewing or approving changes. The classification is based on vendor documentation and is not an endorsement.
When was the Claude Code profile last verified?
Sources for this profile were last reviewed on 15 Oct 2026. 3 sources are listed under Sources, and prices and plans can change, so confirm with the vendor.
Where this information comes from.
- Claude Code documentation map Anthropic (code.claude.com) Official docs Accessed 15 Oct 2026
- Claude Code: AI harness overview innFactory AI Third-party Accessed 15 Oct 2026 Third-party summary of Anthropic's documentation, used for access and data-handling details.
Surfaces and features were checked against Anthropic's Claude Code documentation index. Plan pricing, access and data-handling details come from third-party summaries and should be confirmed with Anthropic. Capabilities were not tested.
Stack18 has not tested this product or received anything from the vendor for this profile. Editorial summaries of vendor information are not endorsements. Confirm details with the vendor before buying.
Evaluating Claude Code?
Check the vendor's current pricing and terms, compare it with similar products, or read how others approach the decision.
Logos provided by Logo.dev. Product names and logos are trademarks of their respective owners and appear for identification only; they do not imply endorsement.