opensource/coder
3
0
Fork 0
mirror of https://github.com/coder/coder.git synced 2026-09-24 15:04:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat: add persisted whole-chat summary with background generation (#26657)

Browse Source 
Adds a persisted whole-chat summary that backs the chat summary popover.
A new nullable `chats.summary` column is populated in the background
after a successful root-chat turn and pushed to clients via a new
`chat_summary_change` watch event (distinct from `summary_change`, which
is bound to `last_turn_summary`), so the popover reads `chat.summary`
straight off the loaded `Chat` with no extra query.

This is the data source for the popover and per-chat cost UI built in
#26649; the popover can consume `chat.summary` once this lands (the
field is nullable, so merge order does not matter).

## How it works

- **Generation** runs in the existing successful-turn finalize hook,
detached from the request so the user's turn is never blocked. A cadence
gate generates the first summary after one completed turn, then
regenerates every three turns, using the `chats.summary_generated_at`
freshness marker. Generation reads compaction-aware history, renders it
to a bounded plain-text transcript (short transcripts are skipped), and
asks for a 1-3 sentence summary via structured output. Failures never
clear an existing summary.
- **Staleness** is guarded by `history_version` (mirroring
`last_turn_summary`), so a background write racing a newer turn loses
while worker lifecycle transitions cannot reject a fresh write.
- **Model selection** uses the chat's configured model.

## Deferred to follow-ups

- **Cost accounting**: the `chat_messages.cost_source` discriminator and
summary/title usage recording were removed from this PR so summary
persistence is not blocked by hidden accounting rows advancing
`history_version`. Title usage recording stays on main's
`InsertChatMessages` path.
- **Model override**: deployment-wide summary generation model selection
is split into #26803; the base feature always uses the chat model.

## Notes

- Migration `000540` adds `chats.summary` and
`chats.summary_generated_at`, and recreates `chats_expanded` to expose
the new columns.
- Root chats only; shared viewers pick up the summary on their next
refetch (live watch events are owner-only).

Refs #26649

---------

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
Jaayden Halko
2026-07-23 16:36:23 +01:00
committed by GitHub
co-authored by Cursor
parent 9ce366414e
commit 3c7a1d33e3
33 changed files with 1473 additions and 120 deletions
Download Patch File Download Diff File Expand all files Collapse all files
coderd/database/queries/chats.sql
+45
View File
@@ -35,6 +35,8 @@ chats_expanded AS (
updated_chats.plan_mode,
updated_chats.client_type,
updated_chats.last_turn_summary,
updated_chats.summary,
updated_chats.summary_generated_at,
updated_chats.snapshot_version,
updated_chats.history_version,
updated_chats.queue_version,
@@ -103,6 +105,8 @@ chats_expanded AS (
updated_chats.plan_mode,
updated_chats.client_type,
updated_chats.last_turn_summary,
updated_chats.summary,
updated_chats.summary_generated_at,
updated_chats.snapshot_version,
updated_chats.history_version,
updated_chats.queue_version,
@@ -836,6 +840,8 @@ chats_expanded AS (
inserted_chat.plan_mode,
inserted_chat.client_type,
inserted_chat.last_turn_summary,
inserted_chat.summary,
inserted_chat.summary_generated_at,
inserted_chat.snapshot_version,
inserted_chat.history_version,
inserted_chat.queue_version,
@@ -978,6 +984,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1046,6 +1054,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1112,6 +1122,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1178,6 +1190,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1244,6 +1258,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1330,6 +1346,8 @@ chats_expanded AS (
result_chat.plan_mode,
result_chat.client_type,
result_chat.last_turn_summary,
result_chat.summary,
result_chat.summary_generated_at,
result_chat.snapshot_version,
result_chat.history_version,
result_chat.queue_version,
@@ -1395,6 +1413,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1437,6 +1457,17 @@ WHERE
id = @id::uuid
AND history_version = @expected_history_version::bigint;
-- name: UpdateChatSummary :execrows
-- The history_version fence lets background summary writes ignore worker-only
-- updates while losing to newer message history.
UPDATE chats
SET
summary = sqlc.narg('summary')::text,
summary_generated_at = NOW()
WHERE
id = @id::uuid
AND history_version = @expected_history_version::bigint;
-- name: UpdateChatMCPServerIDs :one
WITH updated_chat AS (
UPDATE
@@ -1478,6 +1509,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1686,6 +1719,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -1963,6 +1998,8 @@ chats_expanded AS (
locked_chat.plan_mode,
locked_chat.client_type,
locked_chat.last_turn_summary,
locked_chat.summary,
locked_chat.summary_generated_at,
locked_chat.snapshot_version,
locked_chat.history_version,
locked_chat.queue_version,
@@ -2025,6 +2062,8 @@ chats_expanded AS (
shared_chat.plan_mode,
shared_chat.client_type,
shared_chat.last_turn_summary,
shared_chat.summary,
shared_chat.summary_generated_at,
shared_chat.snapshot_version,
shared_chat.history_version,
shared_chat.queue_version,
@@ -2700,6 +2739,8 @@ chats_expanded AS (
bumped_chat.plan_mode,
bumped_chat.client_type,
bumped_chat.last_turn_summary,
bumped_chat.summary,
bumped_chat.summary_generated_at,
bumped_chat.snapshot_version,
bumped_chat.history_version,
bumped_chat.queue_version,
@@ -2775,6 +2816,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,
@@ -2840,6 +2883,8 @@ chats_expanded AS (
updated_chat.plan_mode,
updated_chat.client_type,
updated_chat.last_turn_summary,
updated_chat.summary,
updated_chat.summary_generated_at,
updated_chat.snapshot_version,
updated_chat.history_version,
updated_chat.queue_version,