feat(intelligent-eval): platform triggers OpenClaw agent as worker (avoid external IM channel)

OpenClaw cron requires a channel (announce->last fail-closed); webchat is a
Control-UI feature, not an addressable channel, and platform-side static
execution would degrade the intelligent eval into a static evaluation.

Solution (plan C): the platform keeps the scan loop and, when the queue has
a pending task, invokes the headless agent:
  docker exec openclaw-eval openclaw agent --agent main \
    -m agenteval-intelligent-worker --json
--deliver defaults to false, so no cron delivery channel is involved. The
worker skill runs unchanged under the OpenClaw agent (LLM decisions +
evaluator/analyst skills). Verified headless invocation returns ok.

backend/agenteval/web/app.py

@@ -31,12 +31,61 @@ from agenteval.web.routers import (
 from agenteval.web.websocket import ws_manager
 
 
+def _has_pending_task() -> bool:
+    """True if the intelligent-eval task queue has a pending worker task."""
+    from agenteval.intelligent_eval.task_queue import get_next_task
+
+    session = get_session()
+    try:
+        return get_next_task(session) is not None
+    finally:
+        session.close()
+
+
+async def _trigger_intelligent_worker() -> bool:
+    """Trigger OpenClaw's headless agent to run the intelligent-eval worker skill.
+
+    方案③（避免外部 IM channel）：平台只当"触发闹钟"——有 pending 任务时，
+    用 `docker exec openclaw-eval openclaw agent --agent main -m
+    "agenteval-intelligent-worker"` 唤醒 agent，worker skill 由 OpenClaw agent
+    自主执行（决策/建会话/真实对话/上报，逻辑零改动）。`--deliver` 默认 false
+    因此不经 cron delivery channel。
+
+    Returns:
+        True 若确实触发了 agent（存在 pending 任务）。
+    """
+    if not _has_pending_task():
+        return False
+
+    import subprocess
+
+    _logger = logging.getLogger("agenteval")
+    try:
+        proc = await asyncio.to_thread(
+            subprocess.run,
+            [
+                "docker", "exec", "openclaw-eval", "openclaw", "agent",
+                "--agent", "main", "-m", "agenteval-intelligent-worker", "--json",
+            ],
+            capture_output=True,
+            text=True,
+            timeout=300,
+        )
+        _logger.info("Worker 触发完成 exit=%s", proc.returncode)
+        if proc.returncode != 0:
+            _logger.warning("Worker 触发 stderr: %s", proc.stderr[-300:])
+    except Exception as exc:
+        _logger.warning("Worker 触发失败（忽略）: %s", exc)
+    return True
+
+
 async def _intelligent_eval_scan_loop() -> None:
-    """Scan executing intelligent evals and enqueue worker tasks every minute.
+    """Scan executing intelligent evals, enqueue tasks, and trigger the worker.
 
     v1.1.0 缺陷修复：`scan_and_enqueue_tasks` 此前没有调度点，OpenClaw Worker
-    每分钟唤醒却永远取不到任务。平台启动后每 60s 扫描一次 executing 的评估，
-    与 Worker 的每分钟唤醒对齐。失败不阻断（下次循环继续）。
+    每分钟唤醒却永远取不到任务。平台启动后每 60s：① 扫描 executing 的评估入队；
+    ② 若有 pending 任务则触发 OpenClaw agent 执行 worker skill（方案③，免外部
+    channel）。失败不阻断（下次循环继续）。
     """
     _logger = logging.getLogger("agenteval")
     while True:
@@ -52,6 +101,10 @@ async def _intelligent_eval_scan_loop() -> None:
                 session.close()
         except Exception as exc:
             logging.getLogger("agenteval").warning("智能评估扫描失败（忽略）: %s", exc)
+        try:
+            await _trigger_intelligent_worker()
+        except Exception as exc:
+            logging.getLogger("agenteval").warning("Worker 触发失败（忽略）: %s", exc)
         await asyncio.sleep(60)
 
 

tests/integration/test_intelligent_eval_scan_scheduler.py

@@ -32,3 +32,49 @@ def test_lifespan_starts_scan_loop(monkeypatch):
 
     # The background task runs immediately (before its first 60s sleep).
     assert calls, "scan_and_enqueue_tasks should have been invoked on startup"
+
+
+def test_trigger_worker_skips_when_no_pending(monkeypatch):
+    """No pending task → no docker exec invocation."""
+    import asyncio
+    import subprocess
+    from unittest.mock import MagicMock
+
+    import agenteval.web.app as app_mod
+
+    calls: list = []
+    monkeypatch.setattr(app_mod, "_has_pending_task", lambda: False)
+
+    def fake_run(cmd, **kwargs):
+        calls.append(cmd)
+        return MagicMock(returncode=0, stderr="")
+
+    monkeypatch.setattr(subprocess, "run", fake_run)
+
+    assert asyncio.run(app_mod._trigger_intelligent_worker()) is False
+    assert not calls, "should not invoke docker exec when queue is empty"
+
+
+def test_trigger_worker_calls_docker_exec(monkeypatch):
+    """Pending task present → invoke `docker exec openclaw-eval openclaw agent`."""
+    import asyncio
+    import subprocess
+    from unittest.mock import MagicMock
+
+    import agenteval.web.app as app_mod
+
+    calls: list = []
+    monkeypatch.setattr(app_mod, "_has_pending_task", lambda: True)
+
+    def fake_run(cmd, **kwargs):
+        calls.append(cmd)
+        return MagicMock(returncode=0, stderr="")
+
+    monkeypatch.setattr(subprocess, "run", fake_run)
+
+    assert asyncio.run(app_mod._trigger_intelligent_worker()) is True
+    assert calls, "docker exec should be invoked"
+    joined = " ".join(calls[0])
+    assert "docker" in joined and "openclaw" in joined
+    assert "agenteval-intelligent-worker" in joined
+    assert "--agent" in joined and "main" in joined