gut: remove dead thinking and reasoning session settings#383
Merged
alexey-pelykh merged 3 commits intomainfrom Mar 8, 2026
Merged
gut: remove dead thinking and reasoning session settings#383alexey-pelykh merged 3 commits intomainfrom
alexey-pelykh merged 3 commits intomainfrom
Conversation
auto-merge was automatically disabled
March 8, 2026 01:07
Pull request was closed
166dce3 to
44b9b86
Compare
Remove thinkingLevel and reasoningLevel from all UI surfaces, gateway schema, patch endpoints, and native apps. These settings were resolved and stored per-session but never reached CLI runtimes — AgentRuntime does not forward them, and all four runtimes discard reasoning events. Layers cleaned: - Gateway: schema validation, server methods, sessions-patch logic - Web UI: sessions table, chat view, controllers, types - TUI: /think and /reasoning commands, session types, footer display - CLI: sessions-table columns - macOS: MenuSessionsInjector, SessionActions, SessionData, GatewayModels - Android: ChatController, ChatComposer, ChatSheetContent, NodeRuntime, MainViewModel, TalkModeManager - Shared framework: GatewayModels, ChatModels - Telegram: reasoning-lane-coordinator (deleted), bot-message-dispatch - Matrix: reasoning reply helpers - Auto-reply: directive handling, session resolution, status, commands verboseLevel is preserved everywhere — it gates post-processing delivery and works correctly with AgentRuntime. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…g references (#373) Complete the removal of thinking/reasoning session settings by gutting core type definitions (ThinkLevel, ReasoningLevel, ThinkingContent), Zod schemas, command logic, subagent spawn overrides, heartbeat includeReasoning, ACP translator, and associated tests/docs. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
44b9b86 to
f701c47
Compare
2 tasks
alexey-pelykh
added a commit
that referenced
this pull request
Apr 26, 2026
…2582) Process Cat A cluster A1 of the v2026.3.22 upstream sync backlog: 31 fork files under apps/android/ requiring per-file disposition. Disposition outcome: | Action | Count | |--------|------:| | CHERRY-PICK applied (rebrand transform → fork content changes) | 23 | | CHERRY-PICK no-op (already in sync after rebrand transform) | 4 | | EXCLUDE-DIVERGENT (registered in hq/upstream/disposition.tsv) | 4 | | Total | 31 | EXCLUDE-DIVERGENT classifications: - chat/ChatController.kt + chat/ChatModels.kt — gut #383 removed thinkingLevel/reasoningLevel session settings on fork; upstream still references them. - NodeForegroundService.kt — upstream uses NodeApp.peekRuntime() (nullable) but fork's gutted NodeApp exposes non-null `runtime` only. Reclassified during execution per per-file inspection. - ui/PostOnboardingTabs.kt — upstream calls MainViewModel.refreshHomeCanvasOverviewIfConnected() (new API); fork's gutted MainViewModel exposes requestCanvasRehydrate(source) instead. Reclassified during execution. The 2 reclassifications match the 5–10% rate the issue body anticipated. Approach: 1. Read upstream@v2026.3.22 content via `git show v2026.3.22:{path}`. 2. Apply token-based rebrand transforms per hq/upstream/rebrand-tokens.tsv — path-dependent reverse-domain (`ai.openclaw.app` → `org.remoteclaw.android` for 28 files / `org.remoteclaw.app` for 1 outlier file Base64ImageState.kt and 1 outlier test NodeHandlerRobolectricTest.kt) plus generic transforms (`OpenClaw` → `RemoteClaw`, `openclaw` → `remoteclaw`, `OPENCLAW_` → `REMOTECLAW_`, etc.). 3. Cross-class API audit detected 2 regressions (peekRuntime, refreshHomeCanvasOverviewIfConnected) → reverted those files, reclassified to EXCLUDE-DIVERGENT. Validation gates: pnpm check (oxfmt, tsgo, oxlint, no-random-messaging, no-remoteclaw-ai, css-class-drift), rebrand-gate, zombie-import-gate, stub-debt-gate (126 == baseline), throwing-stub-callers-gate, obsolescence-audit-gate (49 == baseline) — all PASS. Note: Fork CI does not build Android (per CLAUDE.md). Local Android Gradle build not verified. Disposition.tsv updates land in /Users/alexey-pelykh/RemoteClaw/hq/ (HQ companion, not in repo). Four EXTRACT entries with `fork-divergent: ... (#2582)` rationale. Closes #2582
alexey-pelykh
added a commit
that referenced
this pull request
Apr 26, 2026
…2582) (#2595) Process Cat A cluster A1 of the v2026.3.22 upstream sync backlog: 31 fork files under apps/android/ requiring per-file disposition. Disposition outcome: | Action | Count | |--------|------:| | CHERRY-PICK applied (rebrand transform → fork content changes) | 23 | | CHERRY-PICK no-op (already in sync after rebrand transform) | 4 | | EXCLUDE-DIVERGENT (registered in hq/upstream/disposition.tsv) | 4 | | Total | 31 | EXCLUDE-DIVERGENT classifications: - chat/ChatController.kt + chat/ChatModels.kt — gut #383 removed thinkingLevel/reasoningLevel session settings on fork; upstream still references them. - NodeForegroundService.kt — upstream uses NodeApp.peekRuntime() (nullable) but fork's gutted NodeApp exposes non-null `runtime` only. Reclassified during execution per per-file inspection. - ui/PostOnboardingTabs.kt — upstream calls MainViewModel.refreshHomeCanvasOverviewIfConnected() (new API); fork's gutted MainViewModel exposes requestCanvasRehydrate(source) instead. Reclassified during execution. The 2 reclassifications match the 5–10% rate the issue body anticipated. Approach: 1. Read upstream@v2026.3.22 content via `git show v2026.3.22:{path}`. 2. Apply token-based rebrand transforms per hq/upstream/rebrand-tokens.tsv — path-dependent reverse-domain (`ai.openclaw.app` → `org.remoteclaw.android` for 28 files / `org.remoteclaw.app` for 1 outlier file Base64ImageState.kt and 1 outlier test NodeHandlerRobolectricTest.kt) plus generic transforms (`OpenClaw` → `RemoteClaw`, `openclaw` → `remoteclaw`, `OPENCLAW_` → `REMOTECLAW_`, etc.). 3. Cross-class API audit detected 2 regressions (peekRuntime, refreshHomeCanvasOverviewIfConnected) → reverted those files, reclassified to EXCLUDE-DIVERGENT. Validation gates: pnpm check (oxfmt, tsgo, oxlint, no-random-messaging, no-remoteclaw-ai, css-class-drift), rebrand-gate, zombie-import-gate, stub-debt-gate (126 == baseline), throwing-stub-callers-gate, obsolescence-audit-gate (49 == baseline) — all PASS. Note: Fork CI does not build Android (per CLAUDE.md). Local Android Gradle build not verified. Disposition.tsv updates land in /Users/alexey-pelykh/RemoteClaw/hq/ (HQ companion, not in repo). Four EXTRACT entries with `fork-divergent: ... (#2582)` rationale. Closes #2582
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Closes #373.
thinkingLevelandreasoningLevelfrom all UI surfaces (web, TUI, macOS, Android), gateway schema/patch endpoints, and auto-reply directive handlingagent-helpers/thinking.ts— entirely dead code since runtimes discard reasoning eventsverboseLevelpreserved everywhereTest plan
pnpm checkpasses (format + typecheck + lint: 0 warnings, 0 errors)pnpm testpasses (93 test files, 864 tests, all green)thinkingLevelfrom session patch/list assertions and schema validation tests🤖 Generated with Claude Code