reduce inline imports in ai client

src/ai_client.py

@@ -14,13 +14,16 @@ during chat creation to avoid massive history bloat.
 # ai_client.py
 import anthropic
 from google import genai
+from google.api_core import exceptions as gac
 from google.genai import types
+from openai import OpenAI
 import asyncio
 import datetime
 import difflib
 import hashlib
 import json
 import os
+from pathlib import Path as _P
 import requests # type: ignore[import-untyped]
 import sys
 import threading
@@ -38,7 +41,8 @@ from src import performance_monitor
 from src import project_manager
 from src.tool_bias import ToolBiasEngine
 from src.models import ToolPreset, BiasProfile, Tool
-from src.gemini_cli_adapter import GeminiCliAdapter as GeminiCliAdapter
+from src.gemini_cli_adapter import GeminiCliAdapter
+from src.tool_presets import ToolPresetManager
 
 _provider: str = "gemini"
 _model: str = "gemini-2.5-flash-lite"
@@ -356,7 +360,6 @@ def _classify_anthropic_error(exc: Exception) -> ProviderError:
 def _classify_gemini_error(exc: Exception) -> ProviderError:
  body = str(exc).lower()
  try:
-  from google.api_core import exceptions as gac
   if isinstance(exc, gac.ResourceExhausted):
    return ProviderError("quota", "gemini", exc)
   if isinstance(exc, gac.TooManyRequests):
@@ -586,7 +589,6 @@ def set_tool_preset(preset_name: Optional[str]) -> None:
   _active_tool_preset = None
  else:
   try:
-   from src.tool_presets import ToolPresetManager
    manager = ToolPresetManager()
    presets = manager.load_all()
    if preset_name in presets:
@@ -617,7 +619,6 @@ def set_bias_profile(profile_name: Optional[str]) -> None:
   _active_bias_profile = None
  else:
   try:
-   from src.tool_presets import ToolPresetManager
    manager = ToolPresetManager()
    profiles = manager.load_all_bias_profiles()
    if profile_name in profiles:
@@ -886,7 +887,6 @@ def _reread_file_items(file_items: list[dict[str, Any]]) -> tuple[list[dict[str,
   if path is None:
    refreshed.append(item)
    continue
-  from pathlib import Path as _P
   p = path if isinstance(path, _P) else _P(path)
   try:
    current_mtime = p.stat().st_mtime
@@ -1033,8 +1033,7 @@ def _estimate_message_tokens(msg: dict[str, Any]) -> int:
      total_chars += len(text)
     inp = block.get("input")
     if isinstance(inp, dict):
-     import json as _json
+     total_chars += len(json.dumps(inp, ensure_ascii=False))
-     total_chars += len(_json.dumps(inp, ensure_ascii=False))
    elif isinstance(block, str):
     total_chars += len(block)
  est = max(1, int(total_chars / _CHARS_PER_TOKEN))
@@ -2082,7 +2081,6 @@ def _send_deepseek(md_content: str, user_message: str, base_dir: str,
 
 def _list_minimax_models(api_key: str) -> list[str]:
  try:
-  from openai import OpenAI
   client = OpenAI(api_key=api_key, base_url="https://api.minimax.io/v1")
   models_list = client.models.list()
   found = [m.id for m in models_list]
@@ -2095,7 +2093,6 @@ def _list_minimax_models(api_key: str) -> list[str]:
 def _ensure_minimax_client() -> None:
  global _minimax_client
  if _minimax_client is None:
-  from openai import OpenAI
   creds = _load_credentials()
   api_key = creds.get("minimax", {}).get("api_key")
   if not api_key:
@@ -2120,7 +2117,6 @@ def _send_minimax(md_content: str, user_message: str, base_dir: str,
   if not api_key:
    raise ValueError("MiniMax API key not found in credentials.toml")
   
-  from openai import OpenAI
   client = OpenAI(api_key=api_key, base_url="https://api.minimax.io/v1")
   
   with _minimax_history_lock:
@@ -2574,7 +2570,6 @@ def run_subagent_summarization(file_path: str, content: str, is_code: bool, outl
    return f"ERROR: DeepSeek summarization failed: {e}"
  elif _provider == "gemini_cli":
   # Using the adapter for a one-off call
-  from src.gemini_cli_adapter import GeminiCliAdapter
   adapter = GeminiCliAdapter(binary_path="gemini")
   resp_data = adapter.send(prompt, model=_model)
   return resp_data.get("text", "")