diff --git a/client.py b/client.py
index 75471219..26a710f5 100644
--- a/client.py
+++ b/client.py
@@ -38,7 +38,7 @@ def convert_model_for_vertex(model: str) -> str:
Vertex AI uses @ to separate model name from version (e.g., claude-sonnet-4-5@20250929)
while the Anthropic API uses - (e.g., claude-sonnet-4-5-20250929).
- Models without a date suffix (e.g., claude-opus-4-6) pass through unchanged.
+ Models without a date suffix (e.g., claude-opus-4-7) pass through unchanged.
Args:
model: Model name in Anthropic format (with hyphens)
@@ -342,8 +342,10 @@ def create_client(
# Uses get_effective_sdk_env() which reads provider settings from the database,
# ensuring UI-configured alternative providers (GLM, Ollama, Kimi, Custom) propagate
# correctly to the Claude CLI subprocess
- from registry import get_effective_sdk_env
+ from registry import get_effective_sdk_env, get_effort_setting
sdk_env = get_effective_sdk_env()
+ effort = get_effort_setting()
+ print(f" - Reasoning effort: {effort}")
# Detect alternative API mode (Ollama, GLM, or Vertex AI)
base_url = sdk_env.get("ANTHROPIC_BASE_URL", "")
@@ -452,6 +454,9 @@ async def pre_compact_hook(
return ClaudeSDKClient(
options=ClaudeAgentOptions(
model=model,
+ # SDK 0.1.61's effort Literal omits "xhigh" but the CLI's
+ # --effort flag accepts it; the SDK forwards the string unchanged.
+ effort=effort, # type: ignore[arg-type]
cli_path=system_cli, # Use system CLI to avoid bundled Bun crash (exit code 3)
system_prompt="You are an expert full-stack developer building a production-quality web application.",
setting_sources=["project"], # Enable skills, commands, and CLAUDE.md from project dir
diff --git a/registry.py b/registry.py
index 37a2c851..6255d431 100644
--- a/registry.py
+++ b/registry.py
@@ -14,7 +14,7 @@
from contextlib import contextmanager
from datetime import datetime
from pathlib import Path
-from typing import Any
+from typing import Any, Literal, cast
from sqlalchemy import Boolean, Column, DateTime, Integer, String, create_engine, text
from sqlalchemy.orm import DeclarativeBase, sessionmaker
@@ -46,14 +46,17 @@ def _migrate_registry_dir() -> None:
# Available models with display names
# To add a new model: add an entry here with {"id": "model-id", "name": "Display Name"}
AVAILABLE_MODELS = [
- {"id": "claude-opus-4-6", "name": "Claude Opus"},
- {"id": "claude-sonnet-4-5-20250929", "name": "Claude Sonnet"},
+ {"id": "claude-opus-4-7", "name": "Claude Opus"},
+ {"id": "claude-sonnet-4-6", "name": "Claude Sonnet"},
]
# Map legacy model IDs to their current replacements.
# Used by get_all_settings() to auto-migrate stale values on first read after upgrade.
LEGACY_MODEL_MAP = {
- "claude-opus-4-5-20251101": "claude-opus-4-6",
+ "claude-opus-4-5-20251101": "claude-opus-4-7",
+ "claude-opus-4-6": "claude-opus-4-7",
+ "claude-sonnet-4-5": "claude-sonnet-4-6",
+ "claude-sonnet-4-5-20250929": "claude-sonnet-4-6",
}
# List of valid model IDs (derived from AVAILABLE_MODELS)
@@ -65,7 +68,15 @@ def _migrate_registry_dir() -> None:
_env_default_model = os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL")
if _env_default_model is not None:
_env_default_model = _env_default_model.strip()
-DEFAULT_MODEL = _env_default_model or "claude-opus-4-6"
+# Auto-remap stale env-provided values (e.g. user's .env still pins 4.6)
+if _env_default_model and _env_default_model in LEGACY_MODEL_MAP:
+ logging.getLogger(__name__).warning(
+ "ANTHROPIC_DEFAULT_OPUS_MODEL=%s is legacy; remapping to %s. "
+ "Update your .env to silence this warning.",
+ _env_default_model, LEGACY_MODEL_MAP[_env_default_model],
+ )
+ _env_default_model = LEGACY_MODEL_MAP[_env_default_model]
+DEFAULT_MODEL = _env_default_model or "claude-opus-4-7"
# Ensure env-provided DEFAULT_MODEL is in VALID_MODELS for validation consistency
# (idempotent: only adds if missing, doesn't alter AVAILABLE_MODELS semantics)
@@ -671,6 +682,28 @@ def get_setting(key: str, default: str | None = None) -> str | None:
return default
+# Valid Claude Code reasoning/effort levels. Must match the CLI's --effort
+# choices (low, medium, high, xhigh, max) — note: the SDK's Literal type at
+# 0.1.61 omits "xhigh", but the string is forwarded to the CLI as-is and
+# accepted there.
+EffortLevel = Literal["low", "medium", "high", "xhigh", "max"]
+VALID_EFFORT_LEVELS: tuple[EffortLevel, ...] = ("low", "medium", "high", "xhigh", "max")
+DEFAULT_EFFORT: EffortLevel = "xhigh"
+
+
+def get_effort_setting() -> EffortLevel:
+ """
+ Read the global reasoning-effort setting, falling back to ``xhigh``.
+
+ Unknown/invalid stored values are treated as missing so a DB corruption or
+ schema drift can't force the CLI into an unsupported mode.
+ """
+ value = get_setting("effort")
+ if value in VALID_EFFORT_LEVELS:
+ return cast(EffortLevel, value)
+ return DEFAULT_EFFORT
+
+
def set_setting(key: str, value: str) -> None:
"""
Set a setting value (creates or updates).
@@ -699,7 +732,7 @@ def get_all_settings() -> dict[str, str]:
"""
Get all settings as a dictionary.
- Automatically migrates legacy model IDs (e.g. claude-opus-4-5-20251101 -> claude-opus-4-6)
+ Automatically migrates legacy model IDs (e.g. claude-opus-4-6 -> claude-opus-4-7)
on first read after upgrade. This is a one-time silent migration.
Returns:
@@ -747,10 +780,10 @@ def get_all_settings() -> dict[str, str]:
"base_url": None,
"requires_auth": False,
"models": [
- {"id": "claude-opus-4-6", "name": "Claude Opus"},
- {"id": "claude-sonnet-4-5-20250929", "name": "Claude Sonnet"},
+ {"id": "claude-opus-4-7", "name": "Claude Opus"},
+ {"id": "claude-sonnet-4-6", "name": "Claude Sonnet"},
],
- "default_model": "claude-opus-4-6",
+ "default_model": "claude-opus-4-7",
},
"kimi": {
"name": "Kimi K2.5 (Moonshot)",
@@ -778,11 +811,11 @@ def get_all_settings() -> dict[str, str]:
"requires_auth": True,
"auth_env_var": "ANTHROPIC_API_KEY",
"models": [
- {"id": "claude-opus-4-6", "name": "Claude Opus"},
- {"id": "claude-sonnet-4-5", "name": "Claude Sonnet"},
+ {"id": "claude-opus-4-7", "name": "Claude Opus"},
+ {"id": "claude-sonnet-4-6", "name": "Claude Sonnet"},
{"id": "claude-haiku-4-5", "name": "Claude Haiku"},
],
- "default_model": "claude-opus-4-6",
+ "default_model": "claude-opus-4-7",
},
"ollama": {
"name": "Ollama (Local)",
diff --git a/server/routers/settings.py b/server/routers/settings.py
index b0224cd5..9ff04acd 100644
--- a/server/routers/settings.py
+++ b/server/routers/settings.py
@@ -26,6 +26,7 @@
AVAILABLE_MODELS,
DEFAULT_MODEL,
get_all_settings,
+ get_effort_setting,
get_setting,
set_setting,
)
@@ -95,6 +96,8 @@ def _parse_bool(value: str | None, default: bool = False) -> bool:
return value.lower() == "true"
+
+
@router.get("", response_model=SettingsResponse)
async def get_settings():
"""Get current global settings."""
@@ -114,6 +117,7 @@ async def get_settings():
playwright_headless=True, # Always headless - embedded browser view replaces desktop windows
batch_size=_parse_int(all_settings.get("batch_size"), 3),
testing_batch_size=_parse_int(all_settings.get("testing_batch_size"), 3),
+ effort=get_effort_setting(),
api_provider=api_provider,
api_base_url=all_settings.get("api_base_url"),
api_has_auth_token=bool(all_settings.get("api_auth_token")),
@@ -142,6 +146,9 @@ async def update_settings(update: SettingsUpdate):
if update.testing_batch_size is not None:
set_setting("testing_batch_size", str(update.testing_batch_size))
+ if update.effort is not None:
+ set_setting("effort", update.effort)
+
# API provider settings
if update.api_provider is not None:
old_provider = get_setting("api_provider", "claude")
@@ -182,6 +189,7 @@ async def update_settings(update: SettingsUpdate):
playwright_headless=True, # Always headless - embedded browser view replaces desktop windows
batch_size=_parse_int(all_settings.get("batch_size"), 3),
testing_batch_size=_parse_int(all_settings.get("testing_batch_size"), 3),
+ effort=get_effort_setting(),
api_provider=api_provider,
api_base_url=all_settings.get("api_base_url"),
api_has_auth_token=bool(all_settings.get("api_auth_token")),
diff --git a/server/schemas.py b/server/schemas.py
index 3619a480..d54f00aa 100644
--- a/server/schemas.py
+++ b/server/schemas.py
@@ -18,7 +18,7 @@
if str(_root) not in sys.path:
sys.path.insert(0, str(_root))
-from registry import DEFAULT_MODEL, VALID_MODELS
+from registry import DEFAULT_MODEL, LEGACY_MODEL_MAP, VALID_MODELS
# ============================================================================
# Project Schemas
@@ -484,6 +484,7 @@ class SettingsResponse(BaseModel):
playwright_headless: bool = True
batch_size: int = 3 # Features per coding agent batch (1-15)
testing_batch_size: int = 3 # Features per testing agent batch (1-15)
+ effort: Literal["low", "medium", "high", "xhigh", "max"] = "xhigh"
api_provider: str = "claude"
api_base_url: str | None = None
api_has_auth_token: bool = False # Never expose actual token
@@ -504,6 +505,7 @@ class SettingsUpdate(BaseModel):
playwright_headless: bool | None = None
batch_size: int | None = None # Features per agent batch (1-15)
testing_batch_size: int | None = None # Features per testing agent batch (1-15)
+ effort: Literal["low", "medium", "high", "xhigh", "max"] | None = None
api_provider: str | None = None
api_base_url: str | None = Field(None, max_length=500)
api_auth_token: str | None = Field(None, max_length=500) # Write-only, never returned
@@ -520,12 +522,16 @@ def validate_api_base_url(cls, v: str | None) -> str | None:
@field_validator('model')
@classmethod
- def validate_model(cls, v: str | None, info) -> str | None: # type: ignore[override]
+ def validate_model(cls, v: str | None, info) -> str | None:
if v is not None:
# Skip VALID_MODELS check when using an alternative API provider
api_provider = info.data.get("api_provider")
if api_provider and api_provider != "claude":
return v
+ # Transparently accept legacy IDs so in-flight clients don't 422
+ # during an upgrade window; LEGACY_MODEL_MAP already covers migration.
+ if v in LEGACY_MODEL_MAP:
+ v = LEGACY_MODEL_MAP[v]
if v not in VALID_MODELS:
raise ValueError(f"Invalid model. Must be one of: {VALID_MODELS}")
return v
diff --git a/server/services/assistant_chat_session.py b/server/services/assistant_chat_session.py
index cff1f6ce..275a900d 100755
--- a/server/services/assistant_chat_session.py
+++ b/server/services/assistant_chat_session.py
@@ -270,8 +270,9 @@ async def start(self) -> AsyncGenerator[dict, None]:
system_cli = shutil.which("claude")
# Build environment overrides for API configuration
- from registry import DEFAULT_MODEL, get_effective_sdk_env
+ from registry import DEFAULT_MODEL, get_effective_sdk_env, get_effort_setting
sdk_env = get_effective_sdk_env()
+ effort = get_effort_setting()
# Determine model from SDK env (provider-aware) or fallback to env/default
model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", DEFAULT_MODEL)
@@ -281,6 +282,7 @@ async def start(self) -> AsyncGenerator[dict, None]:
self.client = ClaudeSDKClient(
options=ClaudeAgentOptions(
model=model,
+ effort=effort, # type: ignore[arg-type] # SDK 0.1.61 Literal omits "xhigh"
cli_path=system_cli,
# System prompt loaded from CLAUDE.md via setting_sources
# This avoids Windows command line length limit (~8191 chars)
diff --git a/server/services/expand_chat_session.py b/server/services/expand_chat_session.py
index 00a0926c..cb576f1d 100644
--- a/server/services/expand_chat_session.py
+++ b/server/services/expand_chat_session.py
@@ -161,8 +161,9 @@ async def start(self) -> AsyncGenerator[dict, None]:
system_prompt = skill_content.replace("$ARGUMENTS", project_path)
# Build environment overrides for API configuration
- from registry import DEFAULT_MODEL, get_effective_sdk_env
+ from registry import DEFAULT_MODEL, get_effective_sdk_env, get_effort_setting
sdk_env = get_effective_sdk_env()
+ effort = get_effort_setting()
# Determine model from SDK env (provider-aware) or fallback to env/default
model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", DEFAULT_MODEL)
@@ -184,6 +185,7 @@ async def start(self) -> AsyncGenerator[dict, None]:
self.client = ClaudeSDKClient(
options=ClaudeAgentOptions(
model=model,
+ effort=effort, # type: ignore[arg-type] # SDK 0.1.61 Literal omits "xhigh"
cli_path=system_cli,
system_prompt=system_prompt,
allowed_tools=[
diff --git a/server/services/spec_chat_session.py b/server/services/spec_chat_session.py
index 9e853c36..377640c4 100644
--- a/server/services/spec_chat_session.py
+++ b/server/services/spec_chat_session.py
@@ -147,8 +147,9 @@ async def start(self) -> AsyncGenerator[dict, None]:
system_cli = shutil.which("claude")
# Build environment overrides for API configuration
- from registry import DEFAULT_MODEL, get_effective_sdk_env
+ from registry import DEFAULT_MODEL, get_effective_sdk_env, get_effort_setting
sdk_env = get_effective_sdk_env()
+ effort = get_effort_setting()
# Determine model from SDK env (provider-aware) or fallback to env/default
model = sdk_env.get("ANTHROPIC_DEFAULT_OPUS_MODEL") or os.getenv("ANTHROPIC_DEFAULT_OPUS_MODEL", DEFAULT_MODEL)
@@ -157,6 +158,7 @@ async def start(self) -> AsyncGenerator[dict, None]:
self.client = ClaudeSDKClient(
options=ClaudeAgentOptions(
model=model,
+ effort=effort, # type: ignore[arg-type] # SDK 0.1.61 Literal omits "xhigh"
cli_path=system_cli,
# System prompt loaded from CLAUDE.md via setting_sources
# Include "user" for global skills and subagents from ~/.claude/
diff --git a/ui/src/components/SettingsModal.tsx b/ui/src/components/SettingsModal.tsx
index 46ecc790..5605f4e6 100644
--- a/ui/src/components/SettingsModal.tsx
+++ b/ui/src/components/SettingsModal.tsx
@@ -2,7 +2,7 @@ import { useState } from 'react'
import { Loader2, AlertCircle, AlertTriangle, Check, Moon, Sun, Eye, EyeOff, ShieldCheck } from 'lucide-react'
import { useSettings, useUpdateSettings, useAvailableModels, useAvailableProviders } from '../hooks/useProjects'
import { useTheme, THEMES } from '../hooks/useTheme'
-import type { ProviderInfo } from '../lib/types'
+import type { EffortLevel, ProviderInfo } from '../lib/types'
import {
Dialog,
DialogContent,
@@ -70,6 +70,12 @@ export function SettingsModal({ isOpen, onClose }: SettingsModalProps) {
}
}
+ const handleEffortChange = (level: EffortLevel) => {
+ if (!updateSettings.isPending) {
+ updateSettings.mutate({ effort: level })
+ }
+ }
+
const handleProviderChange = (providerId: string) => {
if (!updateSettings.isPending) {
updateSettings.mutate({ api_provider: providerId })
@@ -386,6 +392,30 @@ export function SettingsModal({ isOpen, onClose }: SettingsModalProps) {
)}
+ {/* Reasoning Effort */}
+
+
+
+ How deeply Claude thinks before responding. xhigh is recommended for autonomous coding.
+
+
+ {(['low', 'medium', 'high', 'xhigh', 'max'] as EffortLevel[]).map((level) => (
+
+ ))}
+
+
{/* YOLO Mode Toggle */}
diff --git a/ui/src/components/views/SettingsView.tsx b/ui/src/components/views/SettingsView.tsx
index bf8e6e5f..f22ec034 100644
--- a/ui/src/components/views/SettingsView.tsx
+++ b/ui/src/components/views/SettingsView.tsx
@@ -10,7 +10,7 @@ import { useState } from 'react'
import { Loader2, AlertCircle, AlertTriangle, Check, Moon, Sun, Eye, EyeOff, ShieldCheck, Settings } from 'lucide-react'
import { useSettings, useUpdateSettings, useAvailableModels, useAvailableProviders } from '@/hooks/useProjects'
import { useTheme, THEMES } from '@/hooks/useTheme'
-import type { ProviderInfo } from '@/lib/types'
+import type { EffortLevel, ProviderInfo } from '@/lib/types'
import { Switch } from '@/components/ui/switch'
import { Slider } from '@/components/ui/slider'
import { Label } from '@/components/ui/label'
@@ -68,6 +68,12 @@ export function SettingsView() {
}
}
+ const handleEffortChange = (level: EffortLevel) => {
+ if (!updateSettings.isPending) {
+ updateSettings.mutate({ effort: level })
+ }
+ }
+
const handleProviderChange = (providerId: string) => {
if (!updateSettings.isPending) {
updateSettings.mutate({ api_provider: providerId })
@@ -421,6 +427,30 @@ export function SettingsView() {
/>
+ {/* Reasoning Effort */}
+
+
+
+ How deeply Claude thinks before responding. xhigh is recommended for autonomous coding.
+
+
+ {(['low', 'medium', 'high', 'xhigh', 'max'] as EffortLevel[]).map((level) => (
+
+ ))}
+
+
diff --git a/ui/src/hooks/useProjects.ts b/ui/src/hooks/useProjects.ts
index 707f15ad..88da9c0c 100644
--- a/ui/src/hooks/useProjects.ts
+++ b/ui/src/hooks/useProjects.ts
@@ -288,21 +288,22 @@ export function useValidatePath() {
// Default models response for placeholder (until API responds)
const DEFAULT_MODELS: ModelsResponse = {
models: [
- { id: 'claude-opus-4-6', name: 'Claude Opus' },
- { id: 'claude-sonnet-4-5-20250929', name: 'Claude Sonnet' },
+ { id: 'claude-opus-4-7', name: 'Claude Opus' },
+ { id: 'claude-sonnet-4-6', name: 'Claude Sonnet' },
],
- default: 'claude-opus-4-6',
+ default: 'claude-opus-4-7',
}
const DEFAULT_SETTINGS: Settings = {
yolo_mode: false,
- model: 'claude-opus-4-6',
+ model: 'claude-opus-4-7',
glm_mode: false,
ollama_mode: false,
testing_agent_ratio: 1,
playwright_headless: true,
batch_size: 3,
testing_batch_size: 3,
+ effort: 'xhigh',
api_provider: 'claude',
api_base_url: null,
api_has_auth_token: false,
@@ -311,7 +312,7 @@ const DEFAULT_SETTINGS: Settings = {
const DEFAULT_PROVIDERS: ProvidersResponse = {
providers: [
- { id: 'claude', name: 'Claude (Anthropic)', base_url: null, models: DEFAULT_MODELS.models, default_model: 'claude-opus-4-6', requires_auth: false },
+ { id: 'claude', name: 'Claude (Anthropic)', base_url: null, models: DEFAULT_MODELS.models, default_model: 'claude-opus-4-7', requires_auth: false },
],
current: 'claude',
}
diff --git a/ui/src/lib/types.ts b/ui/src/lib/types.ts
index ae961208..b2ff61aa 100644
--- a/ui/src/lib/types.ts
+++ b/ui/src/lib/types.ts
@@ -642,6 +642,8 @@ export interface ProvidersResponse {
current: string
}
+export type EffortLevel = 'low' | 'medium' | 'high' | 'xhigh' | 'max'
+
export interface Settings {
yolo_mode: boolean
model: string
@@ -651,6 +653,7 @@ export interface Settings {
playwright_headless: boolean
batch_size: number // Features per coding agent batch (1-15)
testing_batch_size: number // Features per testing agent batch (1-15)
+ effort: EffortLevel // Reasoning effort passed to Claude Code (--effort)
api_provider: string
api_base_url: string | null
api_has_auth_token: boolean
@@ -663,6 +666,7 @@ export interface SettingsUpdate {
testing_agent_ratio?: number
batch_size?: number
testing_batch_size?: number
+ effort?: EffortLevel
api_provider?: string
api_base_url?: string
api_auth_token?: string