perf(ttft): salvage easy-win startup optimizations from #3346 (#3395)

* perf(ttft): dedupe shared tool availability checks

* perf(ttft): short-circuit vision auto-resolution

* perf(ttft): make Claude Code version detection lazy

* perf(ttft): reuse loaded toolsets for skills prompt

---------

Co-authored-by: kshitijk4poor <82637225+kshitijk4poor@users.noreply.github.com>

agent/anthropic_adapter.py

@@ -59,6 +59,7 @@ _OAUTH_ONLY_BETAS = [
 # The version must stay reasonably current — Anthropic rejects OAuth requests
 # when the spoofed user-agent version is too far behind the actual release.
 _CLAUDE_CODE_VERSION_FALLBACK = "2.1.74"
+_claude_code_version_cache: Optional[str] = None
 
 
 def _detect_claude_code_version() -> str:
@@ -86,11 +87,18 @@ def _detect_claude_code_version() -> str:
     return _CLAUDE_CODE_VERSION_FALLBACK
 
 
-_CLAUDE_CODE_VERSION = _detect_claude_code_version()
 _CLAUDE_CODE_SYSTEM_PREFIX = "You are Claude Code, Anthropic's official CLI for Claude."
 _MCP_TOOL_PREFIX = "mcp_"
 
 
+def _get_claude_code_version() -> str:
+    """Lazily detect the installed Claude Code version when OAuth headers need it."""
+    global _claude_code_version_cache
+    if _claude_code_version_cache is None:
+        _claude_code_version_cache = _detect_claude_code_version()
+    return _claude_code_version_cache
+
+
 def _is_oauth_token(key: str) -> bool:
     """Check if the key is an OAuth/setup token (not a regular Console API key).
 
@@ -132,7 +140,7 @@ def build_anthropic_client(api_key: str, base_url: str = None):
         kwargs["auth_token"] = api_key
         kwargs["default_headers"] = {
             "anthropic-beta": ",".join(all_betas),
-            "user-agent": f"claude-cli/{_CLAUDE_CODE_VERSION} (external, cli)",
+            "user-agent": f"claude-cli/{_get_claude_code_version()} (external, cli)",
             "x-app": "cli",
         }
     else:
@@ -241,7 +249,7 @@ def _refresh_oauth_token(creds: Dict[str, Any]) -> Optional[str]:
 
     headers = {
         "Content-Type": "application/json",
-        "User-Agent": f"claude-cli/{_CLAUDE_CODE_VERSION} (external, cli)",
+        "User-Agent": f"claude-cli/{_get_claude_code_version()} (external, cli)",
     }
 
     for endpoint in token_endpoints:

agent/auxiliary_client.py

@@ -1137,7 +1137,13 @@ def resolve_vision_provider_client(
         return "custom", client, final_model
 
     if requested == "auto":
-        for candidate in get_available_vision_backends():
+        ordered = list(_VISION_AUTO_PROVIDER_ORDER)
+        preferred = _preferred_main_vision_provider()
+        if preferred in ordered:
+            ordered.remove(preferred)
+            ordered.insert(0, preferred)
+
+        for candidate in ordered:
             sync_client, default_model = _resolve_strict_vision_backend(candidate)
             if sync_client is not None:
                 return _finalize(candidate, sync_client, default_model)

run_agent.py

@@ -62,7 +62,12 @@ else:
 
 
 # Import our tool system
-from model_tools import get_tool_definitions, handle_function_call, check_toolset_requirements
+from model_tools import (
+    get_tool_definitions,
+    get_toolset_for_tool,
+    handle_function_call,
+    check_toolset_requirements,
+)
 from tools.terminal_tool import cleanup_vm
 from tools.interrupt import set_interrupt as _set_interrupt
 from tools.browser_tool import cleanup_browser
@@ -2520,7 +2525,13 @@ class AIAgent:
 
         has_skills_tools = any(name in self.valid_tool_names for name in ['skills_list', 'skill_view', 'skill_manage'])
         if has_skills_tools:
-            avail_toolsets = {ts for ts, avail in check_toolset_requirements().items() if avail}
+            avail_toolsets = {
+                toolset
+                for toolset in (
+                    get_toolset_for_tool(tool_name) for tool_name in self.valid_tool_names
+                )
+                if toolset
+            }
             skills_prompt = build_skills_system_prompt(
                 available_tools=self.valid_tool_names,
                 available_toolsets=avail_toolsets,

tests/agent/test_auxiliary_client.py

@@ -11,6 +11,7 @@ from agent.auxiliary_client import (
     get_text_auxiliary_client,
     get_vision_auxiliary_client,
     get_available_vision_backends,
+    resolve_vision_provider_client,
     resolve_provider_client,
     auxiliary_max_tokens_param,
     _read_codex_access_token,
@@ -638,6 +639,30 @@ class TestVisionClientFallback:
         assert client.__class__.__name__ == "AnthropicAuxiliaryClient"
         assert model == "claude-haiku-4-5-20251001"
 
+    def test_selected_codex_provider_short_circuits_vision_auto(self, monkeypatch):
+        def fake_load_config():
+            return {"model": {"provider": "openai-codex", "default": "gpt-5.2-codex"}}
+
+        codex_client = MagicMock()
+        with (
+            patch("hermes_cli.config.load_config", fake_load_config),
+            patch("agent.auxiliary_client._try_codex", return_value=(codex_client, "gpt-5.2-codex")) as mock_codex,
+            patch("agent.auxiliary_client._try_openrouter") as mock_openrouter,
+            patch("agent.auxiliary_client._try_nous") as mock_nous,
+            patch("agent.auxiliary_client._try_anthropic") as mock_anthropic,
+            patch("agent.auxiliary_client._try_custom_endpoint") as mock_custom,
+        ):
+            provider, client, model = resolve_vision_provider_client()
+
+        assert provider == "openai-codex"
+        assert client is codex_client
+        assert model == "gpt-5.2-codex"
+        mock_codex.assert_called_once()
+        mock_openrouter.assert_not_called()
+        mock_nous.assert_not_called()
+        mock_anthropic.assert_not_called()
+        mock_custom.assert_not_called()
+
     def test_vision_auto_includes_codex(self, codex_auth_dir):
         """Codex supports vision (gpt-5.3-codex), so auto mode should use it."""
         with patch("agent.auxiliary_client._read_nous_auth", return_value=None), \

tests/test_run_agent.py

@@ -584,6 +584,38 @@ class TestBuildSystemPrompt:
         # Should contain current date info like "Conversation started:"
         assert "Conversation started:" in prompt
 
+    def test_skills_prompt_derives_available_toolsets_from_loaded_tools(self):
+        tools = _make_tool_defs("web_search", "skills_list", "skill_view", "skill_manage")
+        toolset_map = {
+            "web_search": "web",
+            "skills_list": "skills",
+            "skill_view": "skills",
+            "skill_manage": "skills",
+        }
+
+        with (
+            patch("run_agent.get_tool_definitions", return_value=tools),
+            patch(
+                "run_agent.check_toolset_requirements",
+                side_effect=AssertionError("should not re-check toolset requirements"),
+            ),
+            patch("run_agent.get_toolset_for_tool", create=True, side_effect=toolset_map.get),
+            patch("run_agent.build_skills_system_prompt", return_value="SKILLS_PROMPT") as mock_skills,
+            patch("run_agent.OpenAI"),
+        ):
+            agent = AIAgent(
+                api_key="test-k...7890",
+                quiet_mode=True,
+                skip_context_files=True,
+                skip_memory=True,
+            )
+
+            prompt = agent._build_system_prompt()
+
+        assert "SKILLS_PROMPT" in prompt
+        assert mock_skills.call_args.kwargs["available_tools"] == set(toolset_map)
+        assert mock_skills.call_args.kwargs["available_toolsets"] == {"web", "skills"}
+
 
 class TestInvalidateSystemPrompt:
     def test_clears_cache(self, agent):

tests/tools/test_registry.py

@@ -81,6 +81,33 @@ class TestGetDefinitions:
         assert len(defs) == 1
         assert defs[0]["function"]["name"] == "available"
 
+    def test_reuses_shared_check_fn_once_per_call(self):
+        reg = ToolRegistry()
+        calls = {"count": 0}
+
+        def shared_check():
+            calls["count"] += 1
+            return True
+
+        reg.register(
+            name="first",
+            toolset="shared",
+            schema=_make_schema("first"),
+            handler=_dummy_handler,
+            check_fn=shared_check,
+        )
+        reg.register(
+            name="second",
+            toolset="shared",
+            schema=_make_schema("second"),
+            handler=_dummy_handler,
+            check_fn=shared_check,
+        )
+
+        defs = reg.get_definitions({"first", "second"})
+        assert len(defs) == 2
+        assert calls["count"] == 1
+
 
 class TestUnknownToolDispatch:
     def test_returns_error_json(self):

tools/registry.py

@@ -98,19 +98,22 @@ class ToolRegistry:
         are included.
         """
         result = []
+        check_results: Dict[Callable, bool] = {}
         for name in sorted(tool_names):
             entry = self._tools.get(name)
             if not entry:
                 continue
             if entry.check_fn:
+                if entry.check_fn not in check_results:
                     try:
-                    if not entry.check_fn():
-                        if not quiet:
-                            logger.debug("Tool %s unavailable (check failed)", name)
-                        continue
+                        check_results[entry.check_fn] = bool(entry.check_fn())
                     except Exception:
+                        check_results[entry.check_fn] = False
                         if not quiet:
                             logger.debug("Tool %s check raised; skipping", name)
+                if not check_results[entry.check_fn]:
+                    if not quiet:
+                        logger.debug("Tool %s unavailable (check failed)", name)
                     continue
             result.append({"type": "function", "function": entry.schema})
         return result