diff --git a/apps/memos-local-plugin/core/feedback/README.md b/apps/memos-local-plugin/core/feedback/README.md
index 2791557af..5d5301a97 100644
--- a/apps/memos-local-plugin/core/feedback/README.md
+++ b/apps/memos-local-plugin/core/feedback/README.md
@@ -240,7 +240,7 @@ import {
 ## Persistence
 
 * `decision_repairs` — the primary table; see
-  [`../storage/repos/decision-repairs.ts`](../storage/repos/decision-repairs.ts).
+  [`../storage/repos/decision_repairs.ts`](../storage/repos/decision_repairs.ts).
   * `context_hash` — anchor for cooldown lookups and retrieval.
   * `preference`, `anti_pattern` — the guidance the agent will see.
   * `high_value_trace_ids`, `low_value_trace_ids` — evidence JSON
diff --git a/apps/memos-local-plugin/docs/DEMO_TaskCLI_OpenClaw_test.md b/apps/memos-local-plugin/docs/DEMO_TaskCLI_OpenClaw_test.md
index ed4cd8efb..6cecd22c5 100644
--- a/apps/memos-local-plugin/docs/DEMO_TaskCLI_OpenClaw_test.md
+++ b/apps/memos-local-plugin/docs/DEMO_TaskCLI_OpenClaw_test.md
@@ -8,7 +8,7 @@
 > 全程使用 `openclaw agent` 命令行 + 浏览器打开 `http://127.0.0.1:18799`。
 > 共 **9 轮交互 + 1 轮收官三层召回**，全程约 25 分钟。
 >
-> 阅读建议：先读 [`Reflect2Skill_算法设计核心.md`](./Reflect2Skill_算法设计核心.md) 0 节与 2.6 节，再读 [`DEMO_OpenClaw_Python_演示.md`](./DEMO_OpenClaw_Python_演示.md)，再读本文。
+> 阅读建议：先读 [`Reflect2Skill_算法设计核心.md`](./Reflect2Skill_算法设计核心.md) 0 节与 2.6 节，再读本文。
 >
 
 ---