feat: increase batch size limits to 15 and add testing_batch_size setting

Batch size configuration:
- Increase coding agent batch size limit from 1-3 to 1-15
- Increase testing agent batch size limit from 1-5 to 1-15
- Add separate `testing_batch_size` setting (previously only CLI-configurable)
- Pass testing_batch_size through full stack: schema → settings router →
  agent router → process manager → CLI flag

UI changes:
- Replace 3-button batch size selector with range slider (1-15)
- Add new Slider component (ui/src/components/ui/slider.tsx)
- Add "Features per Testing Agent" slider in settings panel
- Add custom slider CSS styling for webkit and mozilla

Updated across: CLAUDE.md, autonomous_agent_demo.py, parallel_orchestrator.py,
server/{schemas,routers,services}, and UI types/hooks/components.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

server/routers/agent.py

@@ -17,11 +17,11 @@ from ..utils.project_helpers import get_project_path as _get_project_path
 from ..utils.validation import validate_project_name
 
 
-def _get_settings_defaults() -> tuple[bool, str, int, bool, int]:
+def _get_settings_defaults() -> tuple[bool, str, int, bool, int, int]:
     """Get defaults from global settings.
 
     Returns:
-        Tuple of (yolo_mode, model, testing_agent_ratio, playwright_headless, batch_size)
+        Tuple of (yolo_mode, model, testing_agent_ratio, playwright_headless, batch_size, testing_batch_size)
     """
     import sys
     root = Path(__file__).parent.parent.parent
@@ -47,7 +47,12 @@ def _get_settings_defaults() -> tuple[bool, str, int, bool, int]:
     except (ValueError, TypeError):
         batch_size = 3
 
-    return yolo_mode, model, testing_agent_ratio, playwright_headless, batch_size
+    try:
+        testing_batch_size = int(settings.get("testing_batch_size", "3"))
+    except (ValueError, TypeError):
+        testing_batch_size = 3
+
+    return yolo_mode, model, testing_agent_ratio, playwright_headless, batch_size, testing_batch_size
 
 
 router = APIRouter(prefix="/api/projects/{project_name}/agent", tags=["agent"])
@@ -96,7 +101,7 @@ async def start_agent(
     manager = get_project_manager(project_name)
 
     # Get defaults from global settings if not provided in request
-    default_yolo, default_model, default_testing_ratio, playwright_headless, default_batch_size = _get_settings_defaults()
+    default_yolo, default_model, default_testing_ratio, playwright_headless, default_batch_size, default_testing_batch_size = _get_settings_defaults()
 
     yolo_mode = request.yolo_mode if request.yolo_mode is not None else default_yolo
     model = request.model if request.model else default_model
@@ -104,6 +109,7 @@ async def start_agent(
     testing_agent_ratio = request.testing_agent_ratio if request.testing_agent_ratio is not None else default_testing_ratio
 
     batch_size = default_batch_size
+    testing_batch_size = default_testing_batch_size
 
     success, message = await manager.start(
         yolo_mode=yolo_mode,
@@ -112,6 +118,7 @@ async def start_agent(
         testing_agent_ratio=testing_agent_ratio,
         playwright_headless=playwright_headless,
         batch_size=batch_size,
+        testing_batch_size=testing_batch_size,
     )
 
     # Notify scheduler of manual start (to prevent auto-stop during scheduled window)

server/routers/settings.py

@@ -113,6 +113,7 @@ async def get_settings():
         testing_agent_ratio=_parse_int(all_settings.get("testing_agent_ratio"), 1),
         playwright_headless=_parse_bool(all_settings.get("playwright_headless"), default=True),
         batch_size=_parse_int(all_settings.get("batch_size"), 3),
+        testing_batch_size=_parse_int(all_settings.get("testing_batch_size"), 3),
         api_provider=api_provider,
         api_base_url=all_settings.get("api_base_url"),
         api_has_auth_token=bool(all_settings.get("api_auth_token")),
@@ -138,6 +139,9 @@ async def update_settings(update: SettingsUpdate):
     if update.batch_size is not None:
         set_setting("batch_size", str(update.batch_size))
 
+    if update.testing_batch_size is not None:
+        set_setting("testing_batch_size", str(update.testing_batch_size))
+
     # API provider settings
     if update.api_provider is not None:
         old_provider = get_setting("api_provider", "claude")
@@ -177,6 +181,7 @@ async def update_settings(update: SettingsUpdate):
         testing_agent_ratio=_parse_int(all_settings.get("testing_agent_ratio"), 1),
         playwright_headless=_parse_bool(all_settings.get("playwright_headless"), default=True),
         batch_size=_parse_int(all_settings.get("batch_size"), 3),
+        testing_batch_size=_parse_int(all_settings.get("testing_batch_size"), 3),
         api_provider=api_provider,
         api_base_url=all_settings.get("api_base_url"),
         api_has_auth_token=bool(all_settings.get("api_auth_token")),

server/schemas.py

@@ -444,7 +444,8 @@ class SettingsResponse(BaseModel):
     ollama_mode: bool = False  # True when api_provider is "ollama"
     testing_agent_ratio: int = 1  # Regression testing agents (0-3)
     playwright_headless: bool = True
-    batch_size: int = 3  # Features per coding agent batch (1-3)
+    batch_size: int = 3  # Features per coding agent batch (1-15)
+    testing_batch_size: int = 3  # Features per testing agent batch (1-15)
     api_provider: str = "claude"
     api_base_url: str | None = None
     api_has_auth_token: bool = False  # Never expose actual token
@@ -463,7 +464,8 @@ class SettingsUpdate(BaseModel):
     model: str | None = None
     testing_agent_ratio: int | None = None  # 0-3
     playwright_headless: bool | None = None
-    batch_size: int | None = None  # Features per agent batch (1-3)
+    batch_size: int | None = None  # Features per agent batch (1-15)
+    testing_batch_size: int | None = None  # Features per testing agent batch (1-15)
     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
@@ -500,8 +502,15 @@ class SettingsUpdate(BaseModel):
     @field_validator('batch_size')
     @classmethod
     def validate_batch_size(cls, v: int | None) -> int | None:
-        if v is not None and (v < 1 or v > 3):
-            raise ValueError("batch_size must be between 1 and 3")
+        if v is not None and (v < 1 or v > 15):
+            raise ValueError("batch_size must be between 1 and 15")
+        return v
+
+    @field_validator('testing_batch_size')
+    @classmethod
+    def validate_testing_batch_size(cls, v: int | None) -> int | None:
+        if v is not None and (v < 1 or v > 15):
+            raise ValueError("testing_batch_size must be between 1 and 15")
         return v
 
 

server/services/process_manager.py

@@ -374,6 +374,7 @@ class AgentProcessManager:
         testing_agent_ratio: int = 1,
         playwright_headless: bool = True,
         batch_size: int = 3,
+        testing_batch_size: int = 3,
     ) -> tuple[bool, str]:
         """
         Start the agent as a subprocess.
@@ -440,6 +441,9 @@ class AgentProcessManager:
         # Add --batch-size flag for multi-feature batching
         cmd.extend(["--batch-size", str(batch_size)])
 
+        # Add --testing-batch-size flag for testing agent batching
+        cmd.extend(["--testing-batch-size", str(testing_batch_size)])
+
         # Apply headless setting to .playwright/cli.config.json so playwright-cli
         # picks it up (the only mechanism it supports for headless control)
         self._apply_playwright_headless(playwright_headless)