fix(sessions): swim-35/A1 — sweep remaining raw store[sessionKey] sites (#414) by ronan-dandelion-cult · Pull Request #174 · karmaterminal/openclaw

ronan-dandelion-cult · 2026-04-19T05:03:36Z

swim-35/A1 — Raw `store[sessionKey]` normalization sweep (#414)

Migrates remaining raw store[sessionKey] accesses inside updateSessionStore callbacks to use the canonical resolveSessionStoreEntry pattern (write to normalizedKey + cleanup legacyKeys[]), continuing the swim-26 work landed in fcd4c36.

Comprehension

~/.openclaw/workspace/openclaw-bootstrap/swims/swim-35-stabilization/comprehension/A1-414-comprehension.md

Pattern (canonical)

// BEFORE
store[sessionKey] = { ...store[sessionKey], ...sessionEntry };

// AFTER
const resolved = resolveSessionStoreEntry({ store, sessionKey });
store[resolved.normalizedKey] = { ...resolved.existing, ...sessionEntry };
for (const legacyKey of resolved.legacyKeys) {
  delete store[legacyKey];
}

Files migrated (18 of 18)

#	File	Sites
1	`src/infra/heartbeat-runner.ts`	5
2	`src/agents/auth-profiles/session-override.ts`	2
3	`src/agents/subagent-registry-helpers.ts`	1
4	`src/agents/live-model-switch.ts`	1
5	`src/agents/command/session-store.ts`	2
6	`src/auto-reply/reply/session-reset-model.ts`	1
7	`src/auto-reply/reply/agent-runner-execution.ts`	1 (delete edge case — also iterates `legacyKeys`)
8	`src/auto-reply/reply/abort-cutoff.runtime.ts`	2
9	`src/auto-reply/reply/get-reply-run.ts`	1
10	`src/auto-reply/reply/model-selection.ts`	1
11	`src/auto-reply/reply/session-updates.ts`	2 (incl. `updateCompactionSessionEntry`)
12	`src/auto-reply/reply/directive-handling.persist.ts`	1
13	`src/auto-reply/reply/session.ts`	1 (regression since `fcd4c36`)
14	`src/auto-reply/reply/body.ts`	2
15	`src/auto-reply/reply/directive-handling.impl.ts`	1
16	`src/config/sessions/session-file.ts`	2
17	`extensions/telegram/src/bot-handlers.runtime.ts`	2
18	`extensions/whatsapp/src/auto-reply/heartbeat-runner.ts`	2

Sites skipped: none. All 18 files had migratable raw sites inside updateSessionStore callbacks. The get-reply-run.ts and bot-handlers.runtime.ts "verify" sites flagged in the comprehension turned out to be straightforward additional callbacks; no legacy-cleanup-loop overlap.

Out-of-scope files touched (flagged for review)

src/config/sessions/store.runtime.ts — added resolveSessionStoreEntry to the re-export block so callers reaching the resolver via the runtime alias get it.
extensions/whatsapp/src/auto-reply/heartbeat-runner.runtime.ts — added resolveSessionStoreEntry to the from "openclaw/plugin-sdk/config-runtime" re-export block (mirrors the runtime alias pattern in core).

Test mock updates (4 files)

Migration introduced calls to resolveSessionStoreEntry inside production code that several tests vi.mock(...) for. Added identity-style resolver stubs ({ normalizedKey: sessionKey, existing: store[sessionKey], legacyKeys: [] }) to the partial mocks. Inline comments in each test explain the rationale.

src/agents/live-model-switch.test.ts (2 mock blocks)
src/agents/subagent-registry.test.ts (1 mock block)
src/auto-reply/reply/directive-handling.model.test.ts (1 mock block)
src/auto-reply/reply/directive-handling.mixed-inline.test.ts (1 mock block)

Validation

Typecheck (pnpm tsgo): PASS (clean exit, ~43s)
Vitest:
- vitest.auto-reply-reply.config.ts — 98 files / 1060 tests PASS
- vitest.agents.config.ts — 365/367 files PASS, 3791/3798 tests PASS. The 2 failing files / 3 failing tests are all in src/agents/bash-tools.exec.background-abort.test.ts (pre-existing timing flake: "Matcher did not succeed in time" — unrelated to migration; no bash-tools files were touched on this branch).
- vitest.infra.config.ts — 207 files / 2359 tests PASS
- vitest.runtime-config.config.ts — 131/132 files PASS, 1040/1041 tests PASS. The 1 failure is src/config/schema.base.generated.test.ts (codegen drift for truncateAfterCompaction / tools.exec.applyPatch.enabled — pre-existing on flesh_beast_figs/20260414-claude base; my changes touch neither schema).
- vitest.extension-telegram.config.ts — 91 files / 1257 tests PASS
- vitest.extension-whatsapp.config.ts — 56 files / 447 tests PASS

Commits (6)

a963ed7 — fix(sessions/config): swim-35/A1 — sweep raw store[sessionKey] sites in config/sessions
51d6f68 — fix(agents): swim-35/A1 — sweep raw store[sessionKey] sites in agents/
f55d7f1 — fix(infra/heartbeat): swim-35/A1 — sweep raw store[sessionKey] sites in heartbeat-runner
963d9b4 — fix(auto-reply/reply): swim-35/A1 — sweep raw store[sessionKey] sites in auto-reply/reply/ family
3100a4a — fix(extensions): swim-35/A1 — sweep raw store[sessionKey] sites in telegram + whatsapp
5fb93c2 — test(sessions): swim-35/A1 — add resolveSessionStoreEntry to vi.mock blocks

Refs: #414

…in config/sessions - session-file.ts: resolveAndPersistSessionFile read+write+update-callback - store.runtime.ts: forward resolveSessionStoreEntry export Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

- auth-profiles/session-override.ts: clearOverride + resolveOverride mem+disk - subagent-registry-helpers.ts: findSessionEntryByKey now uses resolver; persistSubagentSessionTiming cleans up legacy duplicates - live-model-switch.ts: clearLiveModelSwitchPending iterates legacyKeys - command/session-store.ts: updateSessionStoreAfterAgentRun read+merge+write Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

…in heartbeat-runner - updateSessionEntryUpdatedAt: load read + update-callback read+write - updateTaskTimestamps: load+save with resolver; legacy cleanup - post-send dedupe write: load+save with resolver; legacy cleanup Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

… in auto-reply/reply/ family - session.ts: regression site + read narrowing fix (entry now SessionEntry|undefined) - session-updates.ts: persistSessionEntryUpdate central helper + incrementCompactionCount + first-turn snapshot read - agent-runner-execution.ts: corrupted-session reset DELETE EDGE CASE — iterates legacyKeys for both mem and disk - abort-cutoff.runtime.ts: clearAbortCutoff mem+disk - body.ts: aborted-hint clear mem+disk via runtime - directive-handling.impl.ts: post-directive persist mem+disk - directive-handling.persist.ts: queue-mode persist mem+disk - get-reply-run.ts: xhigh→high downshift mem+disk - model-selection.ts: disallowed-model reset via runtime - session-reset-model.ts: applyResetModel mem+disk Part of #414 systematic raw-key normalization sweep. The session.ts fix is also the regression Ronan flagged in the comprehension note (a new site introduced after fcd4c36 swim-26 sweep). Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

…legram + whatsapp - telegram/bot-handlers.runtime.ts: model-selection update-callback now resolver-based - whatsapp/heartbeat-runner.ts: sessionId persist (load+save + update-callback) both resolver-based - whatsapp/heartbeat-runner.runtime.ts: forward resolveSessionStoreEntry export Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

elliott-dandelion-cult

APPROVED — A1 sweep is clean and mechanical. 57 resolveSessionStoreEntry call-sites across 20 files, all following the canonical pattern (resolve → read existing → write to normalizedKey → delete legacyKeys). Edge cases in agent-runner-execution.ts and get-reply-run.ts correctly preserve the delete + legacy-cleanup semantics. No scheduler/continuation paths touched. pnpm tsgo clean per Ronan. 🌻

…blocks Migration in swim-35/A1 added resolveSessionStoreEntry calls inside several files whose tests partial-mock the config/sessions modules. Add identity-style resolver stubs (normalizedKey=sessionKey, existing=store[sessionKey], no legacyKeys) to keep the existing test session-key fixtures behavior-equivalent. The real resolver applies session-key normalization; these tests' fixtures are already in canonical form, so the stub is observationally identical for them. Updates: - src/agents/live-model-switch.test.ts (2 mock blocks) - src/agents/subagent-registry.test.ts (1 mock block) - src/auto-reply/reply/directive-handling.model.test.ts (1 mock block) - src/auto-reply/reply/directive-handling.mixed-inline.test.ts (1 mock block) Refs: #414 Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

cael-dandelion-cult

🩸 Cross-sign on A2-decoupling scope: confirmed zero overlap with continuation/scheduler paths. Diff touches session-store + reply-pipeline + heartbeat-runner across 24 files; auto-reply/continuation/ (where continue-delegate.ts + scheduler.ts + context-pressure.ts live) is untouched. resolveSessionStoreEntry sweep is mechanical and read/write-path only. Base flesh_beast_figs/20260414-claude correct. A2 work proceeds independently. 🩸

…le + status restoration Updates RFC docs/design/continue-work-signal-v2.md to reflect the totality of changes since 107ca2b (the prior RFC edit) plus the two ship-gate PRs about to land: - §4.3: document session provider/model threading through volitional compaction (openclaw#191 / bootstrap#639). Three coupled defects: root cause, caller-honesty (phantom-counter), visibility (`unknown_model` classifier + `isLegitSkipReason` helper + `log.warn` on resolve-with-fallback + scope-aware `authProfileId`). - §6.1: add `[context-pressure:noop]` log anchor with reason taxonomy (window-zero / below-threshold / band-dedup); document the bootstrap#580 investigation cycle (`:reach`/`:skip` instrumentation, root cause = sentinel collision on band 0, fix = -1 sentinel). - §6.3: clarify Discord/agent path through src/auto-reply/status.ts was reconnected at openclaw#187 + tested at #188 (the line had been silently dropped in an earlier refactor); note `volitional: N` is honest only after #191. - §6.4: replace 'instrumentation is not currently in place' note with status of distinguishing-instrumentation work (openclaw#164/171/172/173). - Appendix C.1: add 'Closed failure modes' table — phantom-counter, hedge-timer ref leak, band-0 dedup, precondition-skip blindness, Copilot summarization headers, dist-bundle satellite chunks, subagent-announce runtime path mismatch. - Appendix D.2: add evidence-location rows for the new file paths (volitional threading sites; armHedgeTimer; status renderer; request-compaction-tool tests; context-pressure noop sites; agent-runner runtime promotion; subagent-announce co-location; F-NOISE scheduler test). - Header: bump test count (~180 across 13 files, was '172 across 8') to reflect additions in #165, #170, #188, #193. Skip-list (no RFC mention): #174 sessions/config raw-key sweep (internal hygiene); #173 Copilot log-enabled nits (micro-hygiene); 86134af removal of investigation breadcrumbs (cycle is folded into §6.1 narrative). Refs: openclaw#191 head fc3f415 (in-flight, MERGEABLE/UNSTABLE, APPROVED) openclaw#193 head 14483a6 (in-flight, MERGEABLE/UNSTABLE, APPROVED x2) openclaw#187, #188 (merged d787890) openclaw#160, #162, #164, #165, #169, #170, #171, #172, #173, #174 🍆 in 🩲: this is a docs PR; if either #191 or #193 changes shape pre-merge the affected paragraph here will need a one-line touch-up. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

…es (#414) (#174) * fix(sessions/config): swim-35/A1 — sweep raw store[sessionKey] sites in config/sessions - session-file.ts: resolveAndPersistSessionFile read+write+update-callback - store.runtime.ts: forward resolveSessionStoreEntry export Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com> * fix(agents): swim-35/A1 — sweep raw store[sessionKey] sites in agents/ - auth-profiles/session-override.ts: clearOverride + resolveOverride mem+disk - subagent-registry-helpers.ts: findSessionEntryByKey now uses resolver; persistSubagentSessionTiming cleans up legacy duplicates - live-model-switch.ts: clearLiveModelSwitchPending iterates legacyKeys - command/session-store.ts: updateSessionStoreAfterAgentRun read+merge+write Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com> * fix(infra/heartbeat): swim-35/A1 — sweep raw store[sessionKey] sites in heartbeat-runner - updateSessionEntryUpdatedAt: load read + update-callback read+write - updateTaskTimestamps: load+save with resolver; legacy cleanup - post-send dedupe write: load+save with resolver; legacy cleanup Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com> * fix(auto-reply/reply): swim-35/A1 — sweep raw store[sessionKey] sites in auto-reply/reply/ family - session.ts: regression site + read narrowing fix (entry now SessionEntry|undefined) - session-updates.ts: persistSessionEntryUpdate central helper + incrementCompactionCount + first-turn snapshot read - agent-runner-execution.ts: corrupted-session reset DELETE EDGE CASE — iterates legacyKeys for both mem and disk - abort-cutoff.runtime.ts: clearAbortCutoff mem+disk - body.ts: aborted-hint clear mem+disk via runtime - directive-handling.impl.ts: post-directive persist mem+disk - directive-handling.persist.ts: queue-mode persist mem+disk - get-reply-run.ts: xhigh→high downshift mem+disk - model-selection.ts: disallowed-model reset via runtime - session-reset-model.ts: applyResetModel mem+disk Part of #414 systematic raw-key normalization sweep. The session.ts fix is also the regression Ronan flagged in the comprehension note (a new site introduced after fcd4c36 swim-26 sweep). Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com> * fix(extensions): swim-35/A1 — sweep raw store[sessionKey] sites in telegram + whatsapp - telegram/bot-handlers.runtime.ts: model-selection update-callback now resolver-based - whatsapp/heartbeat-runner.ts: sessionId persist (load+save + update-callback) both resolver-based - whatsapp/heartbeat-runner.runtime.ts: forward resolveSessionStoreEntry export Part of #414 systematic raw-key normalization sweep. Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com> * test(sessions): swim-35/A1 — add resolveSessionStoreEntry to vi.mock blocks Migration in swim-35/A1 added resolveSessionStoreEntry calls inside several files whose tests partial-mock the config/sessions modules. Add identity-style resolver stubs (normalizedKey=sessionKey, existing=store[sessionKey], no legacyKeys) to keep the existing test session-key fixtures behavior-equivalent. The real resolver applies session-key normalization; these tests' fixtures are already in canonical form, so the stub is observationally identical for them. Updates: - src/agents/live-model-switch.test.ts (2 mock blocks) - src/agents/subagent-registry.test.ts (1 mock block) - src/auto-reply/reply/directive-handling.model.test.ts (1 mock block) - src/auto-reply/reply/directive-handling.mixed-inline.test.ts (1 mock block) Refs: #414 Co-Authored-By: dandelion cult - ronan 🌊 <karmafeast@gmail.com> --------- Co-authored-by: dandelion cult - ronan 🌊 <karmafeast@gmail.com>

…es (#414) (#174) (#263) * fix(sessions/config): swim-35/A1 — sweep raw store[sessionKey] sites in config/sessions - session-file.ts: resolveAndPersistSessionFile read+write+update-callback - store.runtime.ts: forward resolveSessionStoreEntry export Part of #414 systematic raw-key normalization sweep. * fix(agents): swim-35/A1 — sweep raw store[sessionKey] sites in agents/ - auth-profiles/session-override.ts: clearOverride + resolveOverride mem+disk - subagent-registry-helpers.ts: findSessionEntryByKey now uses resolver; persistSubagentSessionTiming cleans up legacy duplicates - live-model-switch.ts: clearLiveModelSwitchPending iterates legacyKeys - command/session-store.ts: updateSessionStoreAfterAgentRun read+merge+write Part of #414 systematic raw-key normalization sweep. * fix(infra/heartbeat): swim-35/A1 — sweep raw store[sessionKey] sites in heartbeat-runner - updateSessionEntryUpdatedAt: load read + update-callback read+write - updateTaskTimestamps: load+save with resolver; legacy cleanup - post-send dedupe write: load+save with resolver; legacy cleanup Part of #414 systematic raw-key normalization sweep. * fix(auto-reply/reply): swim-35/A1 — sweep raw store[sessionKey] sites in auto-reply/reply/ family - session.ts: regression site + read narrowing fix (entry now SessionEntry|undefined) - session-updates.ts: persistSessionEntryUpdate central helper + incrementCompactionCount + first-turn snapshot read - agent-runner-execution.ts: corrupted-session reset DELETE EDGE CASE — iterates legacyKeys for both mem and disk - abort-cutoff.runtime.ts: clearAbortCutoff mem+disk - body.ts: aborted-hint clear mem+disk via runtime - directive-handling.impl.ts: post-directive persist mem+disk - directive-handling.persist.ts: queue-mode persist mem+disk - get-reply-run.ts: xhigh→high downshift mem+disk - model-selection.ts: disallowed-model reset via runtime - session-reset-model.ts: applyResetModel mem+disk Part of #414 systematic raw-key normalization sweep. The session.ts fix is also the regression Ronan flagged in the comprehension note (a new site introduced after fcd4c36 swim-26 sweep). * fix(extensions): swim-35/A1 — sweep raw store[sessionKey] sites in telegram + whatsapp - telegram/bot-handlers.runtime.ts: model-selection update-callback now resolver-based - whatsapp/heartbeat-runner.ts: sessionId persist (load+save + update-callback) both resolver-based - whatsapp/heartbeat-runner.runtime.ts: forward resolveSessionStoreEntry export Part of #414 systematic raw-key normalization sweep. * test(sessions): swim-35/A1 — add resolveSessionStoreEntry to vi.mock blocks Migration in swim-35/A1 added resolveSessionStoreEntry calls inside several files whose tests partial-mock the config/sessions modules. Add identity-style resolver stubs (normalizedKey=sessionKey, existing=store[sessionKey], no legacyKeys) to keep the existing test session-key fixtures behavior-equivalent. The real resolver applies session-key normalization; these tests' fixtures are already in canonical form, so the stub is observationally identical for them. Updates: - src/agents/live-model-switch.test.ts (2 mock blocks) - src/agents/subagent-registry.test.ts (1 mock block) - src/auto-reply/reply/directive-handling.model.test.ts (1 mock block) - src/auto-reply/reply/directive-handling.mixed-inline.test.ts (1 mock block) Refs: #414 --------- Co-authored-by: Ronan🌊 <ronan.dandelion.cult@hotmail.com>

…es (#414) (#174) (#263) (cherry picked from commit 0da8e31) conflict-resolution: session.ts + session-updates.ts kept squash's resolveSessionStoreEntry form (equivalent semantics, no double-apply)

karmafeast added 5 commits April 18, 2026 22:00

elliott-dandelion-cult approved these changes Apr 19, 2026

View reviewed changes

ronan-dandelion-cult changed the title ~~swim-35/A1: sweep raw store[sessionKey] sites to resolveSessionStoreEntry~~ fix(sessions): swim-35/A1 — sweep remaining raw store[sessionKey] sites (#414) Apr 19, 2026

cael-dandelion-cult approved these changes Apr 19, 2026

View reviewed changes

ronan-dandelion-cult merged commit c3f1f65 into flesh_beast_figs/20260414-claude Apr 19, 2026
3 of 13 checks passed

ronan-dandelion-cult deleted the swim-35-stabilization/ronan-A1 branch April 19, 2026 06:06

ronan-dandelion-cult mentioned this pull request Apr 19, 2026

fix(#639): thread session provider/model through volitional compaction #191

Merged

ronan-dandelion-cult mentioned this pull request Apr 19, 2026

docs(rfc/continuation): fold #191 / #193 / #580-cycle / status restoration into RFC v2 #194

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(sessions): swim-35/A1 — sweep remaining raw store[sessionKey] sites (#414)#174

fix(sessions): swim-35/A1 — sweep remaining raw store[sessionKey] sites (#414)#174
ronan-dandelion-cult merged 6 commits intoflesh_beast_figs/20260414-claudefrom
swim-35-stabilization/ronan-A1

ronan-dandelion-cult commented Apr 19, 2026 •

edited

Loading

Uh oh!

elliott-dandelion-cult left a comment

Uh oh!

cael-dandelion-cult left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

ronan-dandelion-cult commented Apr 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

swim-35/A1 — Raw store[sessionKey] normalization sweep (#414)

Comprehension

Pattern (canonical)

Files migrated (18 of 18)

Out-of-scope files touched (flagged for review)

Test mock updates (4 files)

Validation

Commits (6)

Uh oh!

elliott-dandelion-cult left a comment

Choose a reason for hiding this comment

Uh oh!

cael-dandelion-cult left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

ronan-dandelion-cult commented Apr 19, 2026 •

edited

Loading

swim-35/A1 — Raw `store[sessionKey]` normalization sweep (#414)