diff --git a/apps/memos-local-plugin/adapters/hermes/memos_provider/__init__.py b/apps/memos-local-plugin/adapters/hermes/memos_provider/__init__.py
index fd9d0d827..a67392ca5 100644
--- a/apps/memos-local-plugin/adapters/hermes/memos_provider/__init__.py
+++ b/apps/memos-local-plugin/adapters/hermes/memos_provider/__init__.py
@@ -51,6 +51,7 @@
 import re
 import sys
 import threading
+import weakref
 import time
 
 from pathlib import Path
@@ -1509,6 +1510,21 @@ def on_session_end(self, messages: list[dict[str, Any]]) -> None:  # type: ignor
         with contextlib.suppress(Exception):
             self._bridge.request("session.close", {"sessionId": self._session_id})
 
+    def __del__(self) -> None:
+        # Safety net — if shutdown() was never called (e.g. caller forgot,
+        # Hermes agent routed model change with self.agent = None), clean
+        # up the bridge subprocess and keepalive thread on GC.
+        if self._bridge is not None or (
+            self._bridge_keepalive_thread is not None
+            and self._bridge_keepalive_thread.is_alive()
+        ):
+            logger.warning(
+                "MemOS: __del__ cleaning up leaked provider "
+                "— shutdown() was never called"
+            )
+            with contextlib.suppress(Exception):
+                self.shutdown()
+
     def shutdown(self) -> None:  # type: ignore[override]
         self._bridge_keepalive_stop.set()
         if self._bridge_keepalive_thread and self._bridge_keepalive_thread.is_alive():
@@ -1790,20 +1806,37 @@ def _start_bridge_keepalive(self) -> None:
             return
         self._bridge_keepalive_stop.clear()
 
+        _self_ref = weakref.ref(self)
+
         def _run() -> None:
-            while not self._bridge_keepalive_stop.wait(5.0):
-                if not self._ensure_bridge(self._session_id, timeout=10.0):
+            while True:
+                # Stop signal set (e.g. shutdown called by another thread).
+                # When self is garbage-collected the weakref resolves to None
+                # and we exit gracefully instead of keeping the thread + bridge
+                # subprocess alive forever.
+                provider = _self_ref()
+                if provider is None:
+                    break
+                if provider._bridge_keepalive_stop.wait(5.0):
+                    break
+                if not provider._ensure_bridge(provider._session_id, timeout=10.0):
                     continue
                 try:
-                    assert self._bridge is not None
-                    self._bridge.request("core.health", {}, timeout=10.0)
+                    assert provider._bridge is not None
+                    provider._bridge.request("core.health", {}, timeout=10.0)
                 except Exception as err:
-                    if self._is_transport_closed(err):
-                        logger.info("MemOS: bridge keepalive reconnecting after transport close")
+                    if provider._is_transport_closed(err):
+                        logger.info(
+                            "MemOS: bridge keepalive reconnecting after transport close"
+                        )
                         with contextlib.suppress(Exception):
-                            self._reconnect_bridge(self._session_id, timeout=10.0)
+                            provider._reconnect_bridge(
+                                provider._session_id, timeout=10.0
+                            )
                     else:
-                        logger.debug("MemOS: bridge keepalive failed — %s", err)
+                        logger.debug(
+                            "MemOS: bridge keepalive failed — %s", err
+                        )
 
         self._bridge_keepalive_thread = threading.Thread(
             target=_run,
diff --git a/apps/memos-local-plugin/core/memory/l2/l2.ts b/apps/memos-local-plugin/core/memory/l2/l2.ts
index 9c6685409..903502b32 100644
--- a/apps/memos-local-plugin/core/memory/l2/l2.ts
+++ b/apps/memos-local-plugin/core/memory/l2/l2.ts
@@ -34,7 +34,7 @@ import { L2_INDUCTION_PROMPT } from "../../llm/prompts/l2-induction.js";
 import { associateTraces } from "./associate.js";
 import { makeCandidatePool } from "./candidate-pool.js";
 import { buildPolicyRow, induceDraft } from "./induce.js";
-import { applyGain, computeGain, smoothGain } from "./gain.js";
+import { applyGain, computeGain, nextStatus, smoothGain } from "./gain.js";
 import { signatureOf } from "./signature.js";
 import { tracePolicySimilarity } from "./similarity.js";
 import type {
@@ -455,6 +455,47 @@ export async function runL2(
     timings.gain = Date.now() - t0;
   }
 
+  // ─── Step 5: Re-evaluate untouched candidates ────────────────────────
+  // A candidate policy that was induced in a previous episode but no longer
+  // matches any trace will never enter `touched` and therefore never have
+  // `nextStatus()` run against it. Without this sweep, it stays `candidate`
+  // forever even though its stored gain/support already satisfy the
+  // promotion thresholds.
+  {
+    const untouchedCandidates = repos.policies.list({ status: "candidate" });
+    for (const policy of untouchedCandidates) {
+      if (touched.has(policy.id)) continue; // already handled in Step 4
+      const next = nextStatus({
+        currentStatus: policy.status,
+        support: policy.support,
+        gain: policy.gain,
+        thresholds,
+      });
+      if (next !== policy.status) {
+        repos.policies.updateStats(policy.id, {
+          support: policy.support,
+          gain: policy.gain,
+          status: next,
+          updatedAt: input.now ?? Date.now(),
+        });
+        emit(bus, {
+          kind: "l2.policy.updated",
+          episodeId: input.episodeId,
+          policyId: policy.id,
+          status: next,
+          support: policy.support,
+          gain: policy.gain,
+        });
+        log.info("run.recheck_candidate_promoted", {
+          policyId: policy.id,
+          status: next,
+          support: policy.support,
+          gain: policy.gain,
+        });
+      }
+    }
+  }
+
   timings.persist = 0; // reserved for future split
   const completedAt = Date.now();
   timings.total = completedAt - startedAt;
diff --git a/apps/memos-local-plugin/core/pipeline/orchestrator.ts b/apps/memos-local-plugin/core/pipeline/orchestrator.ts
index 75dc7e244..1e57b4802 100644
--- a/apps/memos-local-plugin/core/pipeline/orchestrator.ts
+++ b/apps/memos-local-plugin/core/pipeline/orchestrator.ts
@@ -1367,6 +1367,7 @@ export function createPipeline(deps: PipelineDeps): PipelineHandle {
     await subs.l3.drain();
     await nextTick();
     await subs.skills.flush();
+    await subs.skills.lifecycleTick();
     await subs.feedback.flush();
     await embeddingRetryWorker.flush();
   }
diff --git a/apps/memos-local-plugin/core/skill/lifecycle.ts b/apps/memos-local-plugin/core/skill/lifecycle.ts
index 49f43bb02..8a8e6daa1 100644
--- a/apps/memos-local-plugin/core/skill/lifecycle.ts
+++ b/apps/memos-local-plugin/core/skill/lifecycle.ts
@@ -201,6 +201,22 @@ export function shouldArchiveIdle(
   return skill.eta < cfg.minEtaForRetrieval;
 }
 
+/**
+ * Decide whether a non-repair candidate skill should skip trial
+ * and become active immediately.  `eta` already encodes the
+ * gain / support of the source policies (initialised at
+ * crystallisation and kept current by reward drift), so we
+ * don't need a separate gain / support gate.
+ */
+export function shouldPromoteCandidate(
+  skill: SkillRow,
+  cfg: SkillConfig,
+): boolean {
+  if (skill.status !== "candidate") return false;
+  if (skill.repairOrigin) return false;
+  return skill.eta >= cfg.minEtaForRetrieval;
+}
+
 function clamp01(n: number): number {
   if (!Number.isFinite(n)) return 0;
   if (n < 0) return 0;
diff --git a/apps/memos-local-plugin/core/skill/subscriber.ts b/apps/memos-local-plugin/core/skill/subscriber.ts
index 52ae6a4f6..664863f92 100644
--- a/apps/memos-local-plugin/core/skill/subscriber.ts
+++ b/apps/memos-local-plugin/core/skill/subscriber.ts
@@ -207,5 +207,19 @@ export function attachSkillSubscriber(
     }
   }
 
-  return { dispose, runOnce, applyFeedback, flush };
+  /** Periodic lifecycle pass: promote eligible candidate skills to active. */
+  async function lifecycleTick(): Promise<void> {
+    const candidates = deps.repos.skills.list({ status: "candidate", limit: 500 });
+    const minEta = deps.config.minEtaForRetrieval ?? 0.1;
+    for (const s of candidates) {
+      if (s.status !== "candidate") continue;
+      if (s.repairOrigin) continue;
+      if ((s.eta ?? 0) < minEta) continue;
+      deps.repos.skills.setStatus(s.id, "active");
+      log.info("skill.auto_promoted", { skillId: s.id, name: s.name, eta: s.eta });
+      try { deps.bus?.emit({ kind: "skill.status.changed", skillId: s.id, from: "candidate", to: "active", reason: "auto_lifecycle" }); } catch { /* best-effort */ }
+    }
+  }
+
+  return { dispose, runOnce, applyFeedback, flush, lifecycleTick };
 }