Skip to content

feat(server): compact idle Scient Agent threads before the cache expires - #923

Merged
2 commits merged into
mainfrom
claude/idle-compaction-20261113
Nov 14, 2026
Merged

2 commits merged into
mainfrom
claude/idle-compaction-20261113

Conversation

@yaacovcorcos

@yaacovcorcos yaacovcorcos commented Nov 14, 2026 •

Copy link
Copy Markdown
Contributor

Platforms: the same server code on macOS, Windows and Linux; no platform-specific parts. Tested live on macOS only.

What

"Compact idle threads" now works for Scient Agent as it does for Claude.

  • Scient Agent reports the prompt-cache entry each reply confirmed on message_end (feat(rpc): tell hosts when the prompt cache expires scient-agent#59). Desktop turns a one-hour entry into promptCacheExpiresAt on the turn's token usage, the field the Claude path already fills (provider/omp/OmpPromptCache.ts).
  • IdleCompaction / IdleCompactionService and the composer's resume check now accept Scient Agent instances (compactsIdleThreads in contracts).
  • Before the cache expires sends /compact two minutes before the cache lapses on a thread over 200,000 tokens. When you return offers the Compact chip, as for Claude.
  • A five-minute entry carries no expiry, so a short pause never compacts. A cache that never expires reports none.
  • Scient Agent instances get the same "Compact idle threads" override as Claude ones, and the setting text names both.
  • effect-omp-rpc event schemas keep the new promptCache field; before this they dropped it.

Older agents: today's agent (0.1.10) sends no promptCache, so nothing changes until the next agent release.

T3 files change only inside SCIENT-FORK seams.

Merge order: #923 → #924 → #925 → #926, all based on main and queued in that order; each branch sits on the one before it, so its diff shrinks to its own change as they merge.

Tests

  • IdleCompaction.scient.test.ts: compaction two minutes before a one-hour cache the agent reported expires; never for a five-minute cache or an agent that reports none; the per-instance override. contextWindow.scient.test.ts: the Compact-on-return chip for Scient Agent.
  • Locally: IdleCompaction*.test.ts, serverSettings.test.ts, OmpAdapterV2.test.ts, contextWindow.scient.test.ts (308 pass), root pnpm run typecheck, vp fmt --check, vp lint, seam check and provenance.
  • Live (dev app, macOS), trigger proven with a shortened TTL: this profile has only ChatGPT models, which declare no cache lifetime, so the agent correctly sent no entry and no expiry was set. With a test-only edit (never committed) that gave each reply a 4-minute expiry and lowered the 200K floor, an idle thread showed "Context automatically compacted before cache expiry" two minutes before the expiry. A real one-hour run needs an Anthropic sign-in and was not done. Record and screenshots: reviews/scient-agent-next-20261111/build-all-20261113/agentmodes-RECORD.md in the ScientFactory workspace.

🤖 Generated with Claude Code

yaacovcorcos and others added 2 commits November 14, 2026 18:58
Scient Agent now reports the prompt-cache entry each reply confirmed
(tier, touch time, lifetime) on its message_end. Desktop turns a one-hour
entry into promptCacheExpiresAt on the turn's token usage, the field the
Claude path already fills. Idle compaction then works for Scient Agent
as it does for Claude: "Before the cache expires" sends /compact two
minutes before the cache lapses on a thread over 200,000 tokens, and
"When you return" offers the Compact chip once the cache has expired.

A five-minute entry carries no expiry, so a short pause never triggers a
compaction, and a cache that never expires reports none. Scient Agent
instances get the same "Compact idle threads" override as Claude ones,
and the setting text names both. Agents that do not report the entry
keep today's behaviour.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The event schemas dropped promptCache, so idle compaction never saw the
cache expiry in the live app.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider Metric Main baseline This PR Impact PR ceiling
Codex Total thread wire 5.5 KiB 5.5 KiB 0 B (0.0%) 6.8 KiB ✅
Codex Thread snapshot wire 4.2 KiB 4.2 KiB 0 B (0.0%) 4.9 KiB ✅
Codex Live turn WebSocket wire 1.3 KiB 1.3 KiB 0 B (0.0%) 2.0 KiB ✅
Codex Live turn WebSocket decoded 22.3 KiB 22.3 KiB 0 B (0.0%) 24.4 KiB ✅
Codex Live turn messages 1 1 0 (0.0%) 8 ✅
Claude Total thread wire 5.5 KiB 5.5 KiB 0 B (0.0%) 6.8 KiB ✅
Claude Thread snapshot wire 4.2 KiB 4.2 KiB 0 B (0.0%) 4.9 KiB ✅
Claude Live turn WebSocket wire 1.3 KiB 1.3 KiB 0 B (0.0%) 2.0 KiB ✅
Claude Live turn WebSocket decoded 22.7 KiB 22.7 KiB 0 B (0.0%) 24.4 KiB ✅
Claude Live turn messages 1 1 0 (0.0%) 8 ✅

Other scenarios

All 64 scenario measurements match main.

All 64 scenario measurements
Provider Scenario Metric Main baseline This PR Impact PR ceiling
Codex short turn resume WebSocket decoded 12.0 KiB 12.0 KiB 0 B (0.0%) 12.7 KiB ✅
Codex short turn resume WebSocket messages 2 2 0 (0.0%) 2 ✅
Codex long turn running, cold open HTTP decoded 281.7 KiB 281.7 KiB 0 B (0.0%) 302.7 KiB ✅
Codex long turn running, cold open rows 75 75 0 (0.0%) 75 ✅
Codex long turn running, history page HTTP decoded 92.4 KiB 92.4 KiB 0 B (0.0%) 107.4 KiB ✅
Codex long turn running, history page rows 75 75 0 (0.0%) 75 ✅
Codex long turn finished, cold open HTTP decoded 381.5 KiB 381.5 KiB 0 B (0.0%) 410.2 KiB ✅
Codex long turn finished, cold open rows 151 151 0 (0.0%) 151 ✅
Codex long turn finished, socket cold open WebSocket decoded 381.6 KiB 381.6 KiB 0 B (0.0%) 410.2 KiB ✅
Codex long turn finished, socket cold open WebSocket messages 2 2 0 (0.0%) 2 ✅
Codex long turn finished, socket cold open rows 151 151 0 (0.0%) 151 ✅
Codex long turn finished, work page HTTP decoded 186.2 KiB 186.2 KiB 0 B (0.0%) 200.2 KiB ✅
Codex long turn finished, work page rows 150 150 0 (0.0%) 150 ✅
Codex idle thread resume WebSocket decoded 65 B 65 B 0 B (0.0%) 200 B ✅
Codex idle thread resume WebSocket messages 1 1 0 (0.0%) 1 ✅
Codex fleet shell snapshot HTTP decoded 45.1 KiB 45.1 KiB 0 B (0.0%) 53.7 KiB ✅
Codex fleet shell snapshot rows 27 27 0 (0.0%) 27 ✅
Codex fleet shell subscribe WebSocket decoded 65 B 65 B 0 B (0.0%) 200 B ✅
Codex fleet shell subscribe WebSocket messages 1 1 0 (0.0%) 1 ✅
Codex fleet shell resume check HTTP decoded 18 B 18 B 0 B (0.0%) 100 B ✅
Codex fleet shell resume check rows 0 0 0 0 ✅
Codex stale shell reload HTTP decoded 45.1 KiB 45.1 KiB 0 B (0.0%) 53.7 KiB ✅
Codex stale shell reload rows 27 27 0 (0.0%) 27 ✅
Codex stale shell resume WebSocket decoded 140 B 140 B 0 B (0.0%) 400 B ✅
Codex stale shell resume WebSocket messages 2 2 0 (0.0%) 2 ✅
Codex stale todo list cold open HTTP decoded 72.4 KiB 72.4 KiB 0 B (0.0%) 83.0 KiB ✅
Codex stale todo list cold open rows 75 75 0 (0.0%) 75 ✅
Codex stale todo list cold open nodes 2 2 0 (0.0%) 4 ✅
Codex stale todo list cold open subagents 2 2 0 (0.0%) 4 ✅
Codex stale todo list cold open plans 1 1 0 (0.0%) 2 ✅
Codex stale todo list cold open providerTurns 1 1 0 (0.0%) 2 ✅
Codex stale todo list cold open checkpoints 1 1 0 (0.0%) 2 ✅
Claude short turn resume WebSocket decoded 12.2 KiB 12.2 KiB 0 B (0.0%) 12.7 KiB ✅
Claude short turn resume WebSocket messages 2 2 0 (0.0%) 2 ✅
Claude long turn running, cold open HTTP decoded 283.3 KiB 283.3 KiB 0 B (0.0%) 302.7 KiB ✅
Claude long turn running, cold open rows 75 75 0 (0.0%) 75 ✅
Claude long turn running, history page HTTP decoded 93.4 KiB 93.4 KiB 0 B (0.0%) 107.4 KiB ✅
Claude long turn running, history page rows 75 75 0 (0.0%) 75 ✅
Claude long turn finished, cold open HTTP decoded 384.1 KiB 384.1 KiB 0 B (0.0%) 410.2 KiB ✅
Claude long turn finished, cold open rows 151 151 0 (0.0%) 151 ✅
Claude long turn finished, socket cold open WebSocket decoded 384.3 KiB 384.3 KiB 0 B (0.0%) 410.2 KiB ✅
Claude long turn finished, socket cold open WebSocket messages 2 2 0 (0.0%) 2 ✅
Claude long turn finished, socket cold open rows 151 151 0 (0.0%) 151 ✅
Claude long turn finished, work page HTTP decoded 188.1 KiB 188.1 KiB 0 B (0.0%) 200.2 KiB ✅
Claude long turn finished, work page rows 150 150 0 (0.0%) 150 ✅
Claude idle thread resume WebSocket decoded 66 B 66 B 0 B (0.0%) 200 B ✅
Claude idle thread resume WebSocket messages 1 1 0 (0.0%) 1 ✅
Claude fleet shell snapshot HTTP decoded 45.6 KiB 45.6 KiB 0 B (0.0%) 53.7 KiB ✅
Claude fleet shell snapshot rows 27 27 0 (0.0%) 27 ✅
Claude fleet shell subscribe WebSocket decoded 66 B 66 B 0 B (0.0%) 200 B ✅
Claude fleet shell subscribe WebSocket messages 1 1 0 (0.0%) 1 ✅
Claude fleet shell resume check HTTP decoded 18 B 18 B 0 B (0.0%) 100 B ✅
Claude fleet shell resume check rows 0 0 0 0 ✅
Claude stale shell reload HTTP decoded 45.6 KiB 45.6 KiB 0 B (0.0%) 53.7 KiB ✅
Claude stale shell reload rows 27 27 0 (0.0%) 27 ✅
Claude stale shell resume WebSocket decoded 140 B 140 B 0 B (0.0%) 400 B ✅
Claude stale shell resume WebSocket messages 2 2 0 (0.0%) 2 ✅
Claude stale todo list cold open HTTP decoded 73.4 KiB 73.4 KiB 0 B (0.0%) 83.0 KiB ✅
Claude stale todo list cold open rows 75 75 0 (0.0%) 75 ✅
Claude stale todo list cold open nodes 2 2 0 (0.0%) 4 ✅
Claude stale todo list cold open subagents 2 2 0 (0.0%) 4 ✅
Claude stale todo list cold open plans 1 1 0 (0.0%) 2 ✅
Claude stale todo list cold open providerTurns 1 1 0 (0.0%) 2 ✅
Claude stale todo list cold open checkpoints 1 1 0 (0.0%) 2 ✅

Baseline: be772d5 · PR result: 0d9b2c3 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.
Bounded compact HTTP snapshot with live cursor.

  • Codex decoded thread snapshot: 88.2 KiB
  • Claude decoded thread snapshot: 88.4 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

@yaacovcorcos
yaacovcorcos added this pull request to the merge queue Nov 14, 2026
@github-merge-queue
github-merge-queue Bot removed this pull request from the merge queue due to failed status checks Nov 14, 2026
@yaacovcorcos
yaacovcorcos added this pull request to the merge queue Nov 14, 2026
@github-merge-queue
github-merge-queue Bot removed this pull request from the merge queue due to it being already merged Nov 14, 2026
@github-merge-queue github-merge-queue Bot closed this pull request by merging all changes into main in 92a8b80 Nov 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

size:L vouch:trusted Author is trusted by repo permissions or the VOUCHED list.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant