> ## Documentation Index¶
> Fetch the complete documentation index at: https://code.claude.com/docs/llms.txt¶
> Use this file to discover all available pages before exploring further.¶
¶
# Output styles¶
¶
> Adapt Claude Code for uses beyond software engineering¶
¶
Output styles change how Claude responds, not what Claude knows. They modify the system prompt to set role, tone, and output format while keeping core capabilities like running scripts, reading and writing files, and tracking TODOs. Use one when you keep re-prompting for the same voice or format every turn, or when you want Claude to act as something other than a software engineer.¶
¶
For instructions about your project, conventions, or codebase, use [CLAUDE.md](/en/memory) instead.¶
¶
## Built-in output styles¶
¶
Claude Code's **Default** output style is the existing system prompt, designed¶
to help you complete software engineering tasks efficiently.¶
¶
There are twohree additional built-in output styles focused on teaching you the¶
codebase and how Claude operates::¶
¶
* **Proactive**: Claude executes immediately, makes reasonable assumptions¶
instead of pausing for routine decisions, and prefers action over planning.¶
This applies the same guidance as¶
[auto mode](/en/permission-modes#eliminate-prompts-with-auto-mode) without¶
changing your permission mode, so you still see permission prompts before¶
tools run.¶
¶
* **Explanatory**: Provides educational "Insights" in between helping you¶
complete software engineering tasks. Helps you understand implementation¶
choices and codebase patterns.¶
¶
* **Learning**: Collaborative, learn-by-doing mode where Claude will not only¶
share "Insights" while coding, but also ask you to contribute small, strategic¶
pieces of code yourself. Claude Code will add `TODO(human)` markers in your¶
code for you to implement.¶
¶
## How output styles work¶
¶
Output styles directly modify Claude Code's system prompt.¶
¶
* Custom output styles exclude instructions for coding (such as verifying code¶
with tests), unless `keep-coding-instructions` is true.¶
* All output styles have their own custom instructions added to the end of the¶
system prompt.¶
* All output styles trigger reminders for Claude to adhere to the output style¶
instructions during the conversation.¶
¶
Token usage depends on the style. Adding instructions to the system prompt¶
increases input tokens, though prompt caching reduces this cost after the first¶
request in a session. The built-in Explanatory and Learning styles produce¶
longer responses than Default by design, which increases output tokens. For¶
custom styles, output token usage depends on what your instructions tell Claude¶
to produce.¶
¶
## Change your output style¶
¶
Run `/config` and select **Output style** to pick a style from a menu. Your¶
selection is saved to `.claude/settings.local.json` at the¶
[local project level](/en/settings).¶
¶
To set a style without the menu, edit the `outputStyle` field directly in a¶
settings file:¶
¶
```json theme={null}¶
{¶
"outputStyle": "Explanatory"¶
}¶
```¶
¶
Because the output style is set in the system prompt at session start,¶
changes take effect the next time you start a new session. This keeps the system¶
prompt stable throughout a conversation so prompt caching can reduce latency and¶
cost.¶
¶
## Create a custom output style¶
¶
Custom output styles are Markdown files with frontmatter and the text that will¶
be added to the system prompt:¶
¶
```markdown theme={null}¶
---¶
name: My Custom Style¶
description:¶
A brief description of what this style does, to be displayed to the user¶
---¶
¶
# Custom Style Instructions¶
¶
You are an interactive CLI tool that helps users with software engineering¶
tasks. [Your custom instructions here...]¶
¶
## Specific Behaviors¶
¶
[Define how the assistant should behave in this style...]¶
```¶
¶
You can save these files at three levels:¶
¶
* User: `~/.claude/output-styles`¶
* Project: `.claude/output-styles`¶
* Managed policy: `.claude/output-styles` inside the [managed settings directory](/en/settings#settings-files)¶
¶
[Plugins](/en/plugins-reference) can also ship output styles in an `output-styles/` directory.¶
¶
### Frontmatter¶
¶
Output style files support frontmatter for specifying metadata:¶
¶
| Frontmatter | Purpose | Default |¶
| :------------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :---------------------- |¶
| `name` | Name of the output style, if not the file name | Inherits from file name |¶
| `description` | Description of the output style, shown in the `/config` picker | None |¶
| `keep-coding-instructions` | Whether to keep the parts of Claude Code's system prompt related to coding. | false |¶
| `force-for-plugin` | Plugin output styles only: apply this style automatically whenever the plugin is enabled, without requiring users to select it. Overrides the user's `outputStyle` setting. If multiple enabled plugins set this, the first one loaded wins. | false |¶
¶
## Comparisons to related features¶
¶
### Output Styles vs. CLAUDE.md vs. --append-system-prompt¶
¶
Output styles completely "turn off" the parts of Claude Code's default system¶
prompt specific to software engineering. Neither CLAUDE.md nor¶
`--append-system-prompt` edit Claude Code's default system prompt. CLAUDE.md¶
adds the contents as a user message *following* Claude Code's default system¶
prompt. `--append-system-prompt` appends the content to the system prompt.¶
¶
### Output Styles vs. [Agents](/en/sub-agents)¶
¶
Output styles directly affect the main agent loop and only affect the system¶
prompt. Agents are invoked to handle specific tasks and can include additional¶
settings like the model to use, the tools they have available, and some context¶
about when to use the agent.¶
¶
### Output Styles vs. [Skills](/en/skills)¶
¶
Output styles modify how Claude responds (formatting, tone, structure) and are always active once selected. Skills are task-specific prompts that you invoke with `/skill-name` or that Claude loads automatically when relevant. Use output styles for consistent formatting preferences; use skills for reusable workflows and tasks.¶