Synara integrates with the Claude Code CLI — Anthropic’s official command-line agent for AI-assisted software development. When you start a Claude thread in Synara, it spawns theDocumentation Index
Fetch the complete documentation index at: https://mintlify.com/Emanuele-web04/synara/llms.txt
Use this file to discover all available pages before exploring further.
claude binary as a local child process and communicates with it over its agent protocol. Synara manages the session lifecycle, approval flows, and context while Claude handles reasoning, code generation, and tool use. The Claude Code CLI must be installed and authorized on your machine before Synara can start a session.
Prerequisites
- Claude Code CLI installed — follow the Anthropic Claude Code installation guide to get started.
- CLI authorized with a valid Anthropic account (run
claudein your terminal and complete the login flow).
Configuration
Claude Code is configured underproviders.claudeAgent in settings.json. All fields are optional.
Set to
false to disable the Claude Code provider entirely. Synara will skip it during startup discovery and will not allow new Claude threads.The command or full path used to launch the Claude Code CLI. Defaults to
"claude", resolved via your PATH. Use an absolute path if your installation is not on PATH.Extra command-line arguments appended to the
claude invocation when Synara starts a session. Accepts any string up to 4 096 characters. Use this to pass flags that the Claude Code CLI supports — for example, custom configuration file paths or experimental feature flags. Leave empty for the default CLI behaviour.Additional model slugs to display in the model picker alongside Synara’s built-in Claude model list. Useful for preview or beta model access.
Model options
Claude model options let you control thinking depth, effort level, and context window per session.Controls the reasoning effort applied before Claude replies. Available values:
| Value | Description |
|---|---|
low | Minimal reasoning — fastest responses. |
medium | Balanced reasoning and speed. |
high | Default for most flagship models — thorough reasoning. |
xhigh | Extra-high reasoning effort. |
max | Maximum standard effort. |
ultrathink | Extended thinking mode; injects a thinking directive into the prompt. Available on Opus 4.7 and Opus 4.8. |
ultracode | Extended code-focused mode (xhigh + agentic workflows). Available on Fable 5 and flagship Opus models. |
Enables or disables the model’s extended thinking capability. Applicable to models that support a thinking toggle (e.g. Claude Haiku 4.5).
Prefer lower-latency responses. Available on Claude Opus 4.7 and Opus 4.8.
Preferred context window size. Supported values for compatible models:
"200k" (default) and "1m".Built-in model roster
| Slug | Name | Effort options |
|---|---|---|
claude-fable-5 | Claude Fable 5 | low · medium · high · xhigh · max · ultracode |
claude-opus-4-8 | Claude Opus 4.8 | low · medium · high · xhigh · max · ultrathink · ultracode |
claude-opus-4-7 | Claude Opus 4.7 | low · medium · high · xhigh · max · ultrathink · ultracode |
claude-opus-4-6 | Claude Opus 4.6 | low · medium · high · max · ultrathink |
claude-opus-4-5 | Claude Opus 4.5 | low · medium · high |
claude-sonnet-4-6 | Claude Sonnet 4.6 | low · medium · high · max · ultrathink |
claude-haiku-4-5 | Claude Haiku 4.5 | thinking toggle only |
claude-sonnet-4-6.