Enable fact and instruction memories from Chat admin settings without agents or actions
I maintainer di solito rispondono entro 1 giorno
@paullizer ci sta già lavorando.
Dal 25/8/2026.
Valutazione
Questa issue non è ancora stata valutata.
Descrizione
Summary
Move the fact/instruction memory control out of Agents & Actions → Actions and surface it in Chat → Chat Experience, so admins can enable memory for the normal chat experience without turning on agents or actions.
This follows the same pattern already used for Tabular Processing: the capability is owned by the tab it actually belongs to, and the Actions pane keeps a read-only dependency note pointing at it.
User Value
Today enable_fact_memory_plugin is only reachable from the Actions tab and is labeled "Enable Fact Memory Action". An admin who never enables agents has no reason to open that tab, so they never discover the switch — even though memory recall already works in plain chat and users can manage entries in Profile.
Framing memory as a chat capability makes it discoverable for the large set of deployments that run plain chat, and removes the false implication that agents are a prerequisite.
Proposed Behavior
enable_fact_memory_pluginremains the single settings key (no new key, no migration).- Admin Settings → Chat → Chat Experience gains a Fact Memory card (
fact-memory-section) containing the live toggle, with copy stating plainly that it works without agents or actions, that users manage entries in Profile, and that the assistant can save or remove memories when asked. - Admin Settings → Agents & Actions → Actions replaces its toggle with a read-only note pointing at Chat, matching the existing
tabular-processing-dependency-notetreatment. - The Chat toggle saves through the main admin settings form;
POST /api/admin/plugins/settingsmoves the key from required to deprecated-optional so it is accepted but no longer written by that endpoint. - With the toggle on and agents off, normal chat also supports explicit memory creation and removal ("remember that I prefer bullet points", "stop calling me Paul") using the existing Semantic Kernel machinery via a small memory-only kernel pass, gated by a cheap intent pre-filter so ordinary turns take no extra model call.
Acceptance Criteria
- The fact memory toggle renders in Chat → Chat Experience and saves through the admin settings form.
- The Actions pane shows a read-only note instead of a toggle, and reflects current state.
- Toggling any other core plugin no longer clobbers
enable_fact_memory_plugin. -
POST /api/admin/plugins/settingsaccepts requests that omit the key, and ignores it when present. - With agents disabled, an explicit user request to remember, change, or forget something persists through
FactMemoryStorein the authorized chat scope. - Standard and streaming chat paths behave consistently, and memory activity surfaces as a chat thought.
- Memory writes never break the chat response — failures degrade silently and are logged.
- Functional regression coverage validates admin placement, save wiring, intent detection, and the memory pass.
- Documentation, docs inventory, and release notes are updated at the configured app version.
Notes
Related to #1153, which covers the broader normal-chat memory creation experience including inferred memory proposal cards with approve/deny/edit. This issue delivers the admin-settings relocation plus the explicit save/update/forget path; the inferred-proposal card UI remains tracked in #1153.
Relevant areas:
application/single_app/templates/admin/_panes/chat-experience.htmlapplication/single_app/templates/admin/_panes/actions.htmlapplication/single_app/admin_settings_nav.pyapplication/single_app/static/js/admin/admin_settings.jsapplication/single_app/route_backend_plugins.pyapplication/single_app/route_frontend_admin_settings.pyapplication/single_app/route_backend_chats.pyapplication/single_app/semantic_kernel_plugins/fact_memory_plugin.pydocs/admin/chat.md,docs/admin/agents-actions.md,docs/_data/features.yml
Hazard found during planning: the Actions toggle currently saves through POST /api/admin/plugins/settings, where enable_fact_memory_plugin is a required field. Removing the input without updating admin_settings.js would make the payload send false, so changing any unrelated core plugin would silently disable fact memory.
- Lingua principale
- Python
- Stelle
- 155
- Fork
- 116
- Merge medio
- 5h 3m
- PR unite (30g)
- 206
Preparare l'ambiente
Avvia il container di sviluppo del progetto nel browser, con il tuo account GitHub.
- Nessun Dockerfile né file Docker Compose
- Ha un modello di pull request
- Leggi la guida per i contributori
Come iniziare
- Leggi tutta la issue e poi la guida ai contributi del progetto.
- Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
- Fai un fork del repository e lavora su un branch.
- Apri una pull request che faccia riferimento al numero della issue.
Altre issue di microsoft/simplechat
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 84/100
microsoft/simplechat#1491 ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 84/100
microsoft/simplechat#1370 ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 5/5 Più di una settimana Idoneità per principianti 8/100
microsoft/simplechat#1770 ·
I maintainer di solito rispondono entro 1 giorno
-
Harden deployer private networking and add optional AMPLS for Commercial and GovernmentForse già presa @paullizer l’ha presa 2 giorni fa. Apertabug enhancement P1 security_improvement
microsoft/simplechat#1765 · 1 assegnatario ·
I maintainer di solito rispondono entro 1 giorno
-
V2 bootstrap takes ~5s: optimize /api/v2/bootstrap backend and first paintForse già presa @paullizer l’ha presa 8 giorni fa. Apertaenhancement P1 python
microsoft/simplechat#1725 · 1 assegnatario ·
I maintainer di solito rispondono entro 1 giorno
Tutte le issue di microsoft/simplechat
Issue simili
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 72/100
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 76/100
omegaup/omegaup#10353 · 1 commento ·
I maintainer di solito rispondono entro 5 giorni
-
bot-triaged module: LrScheduler module: optimizer triaged
Difficoltà 2/5 1-3 ore Idoneità per principianti 70/100
I maintainer di solito rispondono entro 1 giorno
-
documentation glm5 llm-model
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
vllm-project/vllm-ascend#19062 ·
I maintainer di solito rispondono entro 1 giorno
-
Author Page: shuai-zhao-llmApertaauthor-page correction
Difficoltà 2/5 1-3 ore Idoneità per principianti 68/100
acl-org/acl-anthology#10496 · 1 commento ·
I maintainer di solito rispondono entro 1 giorno