fix: bound A2A orphan grace and clear _active_tasks on disconnect

`_orphan_timeout()` was `max(300, A2A_REPLY_TIMEOUT)` with no ceiling, so
an absurd value (1e18) meant the watchdog sweep could never fail an
orphan — the reply window is a floor for the grace, not a licence to
disable the sweep. Cap it at 86400s.

`disconnect()` failed and cleared `_pending`/`_pending_order` but left
`_active_tasks` populated, so a reconnected adapter would keep excluding
dead task ids from the orphan sweep forever. Clear it in the same locked
block.

plugins/platforms/a2a/adapter.py

@@ -33,7 +33,9 @@ from . import protocol, security
 logger = logging.getLogger(__name__)
 
 _DEFAULT_PORT = 9900
-_MIN_ORPHAN_TIMEOUT, _WATCHDOG_INTERVAL = 300, 60  # seconds: orphan grace floor / watchdog period
+# seconds: orphan grace floor / ceiling / watchdog period. The ceiling keeps the sweep
+# meaningful when A2A_REPLY_TIMEOUT is absurd (1e18 would never fail an orphan).
+_MIN_ORPHAN_TIMEOUT, _MAX_ORPHAN_TIMEOUT, _WATCHDOG_INTERVAL = 300, 86400, 60
 _MAX_BODY = 1_048_576  # 1MB max request body — prevents DoS via memory exhaustion
 _SSE_KEEPALIVE = 5  # seconds between SSE keepalive comments
 _DEFAULT_DESCRIPTION = "Hermes Agent — a general-purpose agent reachable over A2A."
@@ -70,8 +72,8 @@ def _reply_timeout() -> float:
 
 
 def _orphan_timeout() -> float:
-    """Orphan grace must never expire before a configured reply window."""
-    return max(float(_MIN_ORPHAN_TIMEOUT), _reply_timeout())
+    """Orphan grace must never expire before a configured reply window, but stays bounded."""
+    return min(float(_MAX_ORPHAN_TIMEOUT), max(float(_MIN_ORPHAN_TIMEOUT), _reply_timeout()))
 
 
 def _default_agent_name() -> str:
@@ -336,6 +338,7 @@ class A2AAdapter(BasePlatformAdapter):
                 self._resolve_locked(tid, protocol.STATE_FAILED, "[agent shutting down]")
             self._pending.clear()
             self._pending_order.clear()
+            self._active_tasks.clear()
 
     def _watchdog_loop(self) -> None:
         """Background thread that fails orphaned tasks (keeps them queryable)."""

tests/plugins/test_a2a_phase23.py

@@ -534,6 +534,16 @@ class TestTaskStore:
         adapter._pop_pending("t-live")
         assert adapter._fail_orphans_once() == ["t-live"]
 
+    def test_orphan_timeout_is_bounded_and_disconnect_clears_active_tasks(self, monkeypatch):
+        from plugins.platforms.a2a import adapter as mod
+        monkeypatch.setenv("A2A_REPLY_TIMEOUT", "1e18")
+        assert mod._orphan_timeout() == mod._MAX_ORPHAN_TIMEOUT
+
+        adapter, _base = _make_live_adapter(monkeypatch)
+        adapter._add_pending("t-live", "c1")
+        asyncio.run(adapter.disconnect())
+        assert adapter._active_tasks == set()
+
     def test_watchdog_cannot_race_local_finalization(self, monkeypatch):
         adapter, _base = _make_live_adapter(monkeypatch)
         rec = adapter.tasks.create("t-live", "c1", "peer")