From ae73bbde274c6e6cb4f51ab8e0c13caf8a22acf9 Mon Sep 17 00:00:00 2001
From: Adam Cross <adam.cross@ragu.ai>
Date: Mon, 20 Jul 2026 23:27:44 +0100
Subject: [PATCH] docs(memory): document agent.enabled opt-in for automatic
 extraction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Documents the breaking change from PR #12886 (merged 2026-05-01):
`memory.agent.enabled: true` is now required for the automatic memory
extraction agent to run. Configs with an `agent` block but no explicit
`enabled: true` log a startup warning and silently skip extraction —
manual memory operations still work.

Changes:
- add `enabled: true` to every `agent:` example (top example, agent-by-id,
  custom agent, complete example, custom-endpoint example)
- document `enabled` as an Agent Configuration Field with the opt-in
  default and a link back to PR #12886
- add an info callout under the top-level example showing the exact
  warning text, and a bullet under Notes summarising the opt-in behaviour

Fixes documentation gap surfaced by BT prod: memory extraction stopped
silently after upgrading past PR #12886 because the docs still showed the
pre-change example.
---
 .../object_structure/memory.mdx               | 37 +++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/content/docs/configuration/librechat_yaml/object_structure/memory.mdx b/content/docs/configuration/librechat_yaml/object_structure/memory.mdx
index 12166c734..08e15bd75 100644
--- a/content/docs/configuration/librechat_yaml/object_structure/memory.mdx
+++ b/content/docs/configuration/librechat_yaml/object_structure/memory.mdx
@@ -19,6 +19,7 @@ memory:
   personalize: true
   messageWindowSize: 5
   agent:
+    enabled: true
     provider: 'openAI'
     model: 'gpt-4'
     instructions: 'You are a helpful assistant that remembers user preferences and context.'
@@ -27,6 +28,22 @@ memory:
       max_tokens: 1000
 ```
 
+<Callout type="info">
+  As of [PR #12886](https://github.com/danny-avila/LibreChat/pull/12886), the memory
+  **agent** must set `enabled: true` for automatic memory extraction to run.
+  If the flag is omitted, LibreChat logs a startup warning and skips automatic
+  extraction. Manual memory operations (create, edit, delete, reference)
+  continue to work regardless of the flag.
+
+  The warning looks like:
+
+  ```text
+  [memory] Agent config detected without explicit `enabled: true`.
+  Automatic memory extraction is now opt-in.
+  Add `memory.agent.enabled: true` to keep automatic extraction on.
+  ```
+</Callout>
+
 ## disabled
 
 <OptionTable
@@ -193,6 +210,7 @@ When you have a pre-configured agent, you can reference it by its ID:
 ```yaml filename="memory / agent (by ID)"
 memory:
   agent:
+    enabled: true
     id: 'memory-agent-001'
 ```
 
@@ -203,6 +221,7 @@ For more control, you can define a complete agent configuration:
 ```yaml filename="memory / agent (custom)"
 memory:
   agent:
+    enabled: true
     provider: 'openAI'
     model: 'gpt-4'
     instructions: 'You are a memory assistant that helps maintain conversation context and user preferences.'
@@ -216,6 +235,21 @@ memory:
 
 When using custom agent configuration, the following fields are available:
 
+**enabled** (required for automatic extraction)
+
+<OptionTable
+  options={[
+    [
+      'enabled',
+      'Boolean',
+      'Explicitly enables the automatic memory-extraction agent. As of [PR #12886](https://github.com/danny-avila/LibreChat/pull/12886), an `agent` block without `enabled: true` logs a warning at startup and does not run automatic extraction. Manual memory operations remain available.',
+      'enabled: true',
+    ],
+  ]}
+/>
+
+**Default:** `false` (opt-in).
+
 **provider** (required)
 
 <OptionTable
@@ -284,6 +318,7 @@ memory:
   personalize: true
   messageWindowSize: 8
   agent:
+    enabled: true
     provider: 'openAI'
     model: 'gpt-4'
     instructions: |
@@ -330,6 +365,7 @@ memory:
   personalize: true
   messageWindowSize: 10
   agent:
+    enabled: true
     provider: 'Custom Memory Endpoint'
     model: 'gpt-4o-mini'
 ```
@@ -345,5 +381,6 @@ memory:
 - Valid keys provide granular control over what information can be stored
 - Custom `instructions` replace default memory handling instructions and should be used with `validKeys`
 - Agent configuration allows customization of memory processing behavior
+- Automatic memory extraction is opt-in: `memory.agent.enabled: true` is required (see [PR #12886](https://github.com/danny-avila/LibreChat/pull/12886)). Manual memory operations remain available even without an enabled agent
 - When disabled, all memory features are turned off regardless of other settings
 - The message window size affects how much recent context is considered for memory updates