fix(agent): append system_prompt to the Claude Code preset instead of replacing

src/coder_eval/agents/claude_code_agent.py

@@ -27,6 +27,7 @@
 # CLI). If this import breaks on an SDK upgrade, the threaded watchdog loses
 # its kill target and timeouts will no longer be enforced at the agent layer.
 from claude_agent_sdk._internal.transport.subprocess_cli import SubprocessCLITransport
+from claude_agent_sdk.types import SystemPromptPreset
 
 from coder_eval.agent import Agent, AgentState
 from coder_eval.agents._logging import PrefixedAdapter, log_raw_sdk_event
@@ -1173,6 +1174,14 @@ def _build_claude_query(
         if "ToolSearch" not in disallowed_tools:
             disallowed_tools.append("ToolSearch")
 
+        # A plain-string system_prompt would REPLACE Claude Code's default system
+        # prompt, dropping its behavioral guidance (parallel tool-call batching,
+        # conciseness). Always keep the default via the SDK preset and append the
+        # configured prompt after it.
+        system_prompt: SystemPromptPreset | None = None
+        if self.config.system_prompt is not None:
+            system_prompt = SystemPromptPreset(type="preset", preset="claude_code", append=self.config.system_prompt)
+
         # as_posix(), not str(): bash on Windows strips backslashes from unquoted
         # paths, so a redirect like `> D:\foo\bar` ends up writing to "Dfoobar".
         options = ClaudeAgentOptions(
@@ -1192,7 +1201,7 @@ def _build_claude_query(
             # summing per-message values undercounts by 10x+. Without this flag
             # StreamEvents are suppressed by the SDK.
             include_partial_messages=True,
-            system_prompt=self.config.system_prompt,
+            system_prompt=system_prompt,
             setting_sources=self.config.setting_sources if self.config.setting_sources is not None else ["project"],
             resume=self._session_id,
             settings=json.dumps(self.config.claude_settings)

src/coder_eval/models/agent_config.py

@@ -151,7 +151,8 @@ class BaseAgentConfig(BaseModel):
     system_prompt: str | None = Field(
         default=None,
         description=(
-            "Custom system prompt. Replaces the default system prompt. "
+            "Custom system prompt, appended to the agent's default system prompt "
+            "(claude-code: the SDK 'claude_code' preset with append). "
             "Supports inline text or multi-line YAML strings. "
             "Mutually exclusive with system_prompt_file."
         ),

tests/test_agent.py

@@ -377,6 +377,32 @@ async def test_claude_settings_none_default():
     assert captured_options[0].settings is None
 
 
+@pytest.mark.asyncio
+async def test_system_prompt_appends_to_claude_code_preset():
+    """system_prompt keeps the Claude Code default prompt and appends via the SDK preset."""
+    config = parse_agent_config(type=AgentKind.CLAUDE_CODE, system_prompt="You are a coding agent.")
+    agent = ClaudeCodeAgent(config)
+
+    captured_options = await _capture_sdk_options(agent)
+
+    assert captured_options[0].system_prompt == {
+        "type": "preset",
+        "preset": "claude_code",
+        "append": "You are a coding agent.",
+    }
+
+
+@pytest.mark.asyncio
+async def test_system_prompt_none_leaves_sdk_default():
+    """No system_prompt -> ClaudeAgentOptions.system_prompt stays None (SDK default prompt)."""
+    config = parse_agent_config(type=AgentKind.CLAUDE_CODE)
+    agent = ClaudeCodeAgent(config)
+
+    captured_options = await _capture_sdk_options(agent)
+
+    assert captured_options[0].system_prompt is None
+
+
 @pytest.mark.asyncio
 async def test_sdk_options_forwarded_to_sdk():
     """An sdk_options key (e.g. effort) is splatted into ClaudeAgentOptions."""