- composer_history: raise deadline to 10s on Windows for writer thread
- prompts: make memory_guidance tier-order assertion CRLF-agnostic
- gitattributes: enforce LF line endings for include_str!() prompt files
Regression from the v0.8.44 release changes — the writer thread batching
and the updated constitutional tier ordering in memory_guidance.md both
uncovered Windows-only test flakes.
migrate_config_if_needed() copies config.toml on first launch.
Called in run_interactive after config creation. Non-fatal,
never overwrites an existing primary config.
#1981: DelegateCard now shows readable roles (scout, builder,
reviewer, etc.) instead of raw agent_type strings, and uses the
completion summary as the primary detail when available. Raw
agent_ids are truncated to 8 chars as secondary detail.
#834: PlanPromptView now renders the plan explanation and steps
(when update_plan was called) before the action options. Uses
PlanSnapshot from the shared plan state.
#1638: Session picker auto-opens on clean startup when saved
sessions exist and no --resume/--continue/--fresh was specified.
No new TuiOptions field needed — uses existing resume_session_id.
When the model is busy but not actively streaming text (waiting on
tool results, sub-agents, or shell commands), Enter now tries to steer
the message into the current turn rather than silently queueing it.
If steering fails, the message falls back to queuing.
During active streaming, Enter still queues to avoid interrupting
in-flight reasoning. Ctrl+Enter forces Steer in all busy states.
#1600: 'r' keybinding in session picker for inline rename.
Enter inline rename mode, type new title, Enter to confirm, Esc to cancel.
Updates the saved session metadata and refreshes the picker list.
#1787: checklist_write, checklist_update, update_plan now trigger
immediate Work sidebar refresh (previously only todo_write did).
Permanent short-form shim — forwards silently to codewhale.
Six fewer keystrokes, no deprecation warning. Ships alongside
codewhale and the legacy deepseek alias.
- Doctor: add 'State Root' section showing active root, legacy status,
and dual-root detection
- Doctor (text + JSON): use codewhale_config resolver for config path
- deepseek_home_dir(): now resolves to CodeWhale home
- composer_stash: writes to ~/.codewhale/composer_stash.jsonl
- merge_project_config: checks .codewhale/ before .deepseek/
- Text/config paths default to .codewhale/ not .deepseek/
#2011: migrate app state to ~/.codewhale
- Add CodeWhalePaths: codewhale_home(), legacy_deepseek_home(),
resolve_state_dir(), ensure_state_dir() in codewhale-config
- Config: resolve_config_path supports CODEWHALE_CONFIG_PATH env,
default_config_path prefers ~/.codewhale/config.toml
- Project overlay: checks .codewhale/config.toml before .deepseek/
- Sessions: default_sessions_dir uses resolve_state_dir with fallback
- Workspace trust: writes to CodeWhale home via ensure_state_dir
- Init: ensure_deepseek_gitignored adds both .codewhale/ and .deepseek/
- .gitignore: adds .codewhale/
#2010: session artifact hygiene
- /save without path now writes to managed sessions dir instead of cwd
- Boot-time session prune via cleanup_old_sessions (MAX_SESSIONS=50)
- sessions_dir() public accessor for checkpoint path resolution
Fix: load_recent_checkpoint now uses manager.sessions_dir() instead
of hardcoding ~/.deepseek/sessions/checkpoints/
Fixes the macOS test failure on PR #1988 and the contributor-credit
gate from scripts/release/check-versions.sh.
cell_has_detail_target() was matching HistoryCell::Thinking, which
caused the activity footer to append " · ⌥+V raw" to thinking cells
that have no separate raw detail target. The detail-card flow only
exists for tool / sub-agent cells; thinking renders its own raw text
inline. Removing Thinking from the match arm restores the behavior the
existing
activity_footer_hint_surfaces_visible_thinking_without_raw_tool_hint
test asserts.
The CHANGELOG.md 0.8.43 section now credits the 30 contributors added
to README acknowledgements in this cycle, satisfying the README-vs-
CHANGELOG cross-check in check-versions.sh. crates/tui/CHANGELOG.md is
re-synced so the matching guard passes.
Verified locally on macOS:
- cargo fmt --all -- --check : clean
- cargo clippy --workspace --all-targets --all-features --locked -D warnings : clean
- cargo test --workspace --all-features --locked : 41 suites, 0 failed
- ./scripts/release/check-versions.sh : Version state OK
- ./scripts/release/publish-crates.sh dry-run : all 14 crates OK
- cargo build --release --locked : clean
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- npm/codewhale/README.md: remove DeepSeek-first language
- docs/INSTALL.md: scoop install codewhale (not deepseek-tui)
- Wire decision card overlay rendering in main render loop
- Decision cards now appear centered on transcript when active
Before the turn breaks at the thinking-only checkpoint, drain
any sub-agent completions that arrived between the last hold
check and now. If a child finished while we were running the
final status check, surface its sentinel immediately rather
than delaying it to the next turn.
- Thinking cells now qualify as detail targets alongside tools and sub-agents
- Space key on empty composer toggles the focused cell's collapsed state
- Status message confirms expand/collapse action
- Builds on existing collapsed_cells HashSet from mouse context menu
Implements structured choice cards for the v0.8.43 truth-surface tracker.
When Brother Whale needs user input, it surfaces a bordered card with:
- A question prompt
- Numbered options with (default) marker
- Arrow/j/k navigation and 1-9 number-key shortcuts
- Enter to confirm, Esc to cancel
- Decision results surfaced as status messages
The widget compiles and the keyboard routing is wired into the main
event loop. Rendering overlay wire-up follows in the next commit.
When a turn has been running for > 30 seconds, the footer now shows
a classified stall reason suffix:
- 'waiting for model' — API streaming in progress
- 'tools executing' — active tool calls running
- 'sub-agents working' — child sub-agents in flight
- 'compacting context' — context compaction active
- 'background jobs running' — shell tasks executing
- 'waiting — no recent activity' — turn stalled with no classified work
- Show fuller turn ID prefix (16 chars) for disambiguation in task_read/task_cancel
- Replace ambiguous 'X active (Y running)' with clear per-status breakdown
- Add y/Y yank affordances for copying turn ID and full status from Tasks panel
- Add yank hint text in Tasks panel footer
- Separate model auto-routing from Plan/Agent/YOLO TUI modes across all
READMEs and docs/MODES.md
- Introduce Fin as the fast thinking-off deepseek-v4-flash seam for
routing, summaries, RLM child calls, and coordination work
- Document /goal as current session tracking (not a TUI mode), with note
that a future Goal work surface should stay distinct from --model auto
- Extend deprecation shim timeline from 'one release cycle' to
through v0.8.x in REBRAND.md and npm READMEs
- Fix selection_to_text test to expect inline thinking preview
(short completed thinking now renders inline without Ctrl+O
affordance)
DeepSeek native accepts reasoning_effort: high | max, with low/medium
compat-mapping to high and xhigh mapping to max. OpenRouter and Novita
use the OpenAI-standard unified scale: none | minimal | low | medium |
high | xhigh.
Mapping (CodeWhale internal → wire):
off → DeepSeek: thinking disabled, OpenRouter: thinking disabled
low → DeepSeek: high, OpenRouter: low
medium → DeepSeek: high, OpenRouter: medium
high → DeepSeek: high, OpenRouter: high
max → DeepSeek: max, OpenRouter: xhigh
Previously the low/medium/high arm sent 'high' for all providers,
and the max arm sent 'max' for all providers — both wrong for
OpenRouter. Split each arm to preserve provider-native semantics.
Sources:
- https://api-docs.deepseek.com/guides/thinking_mode
- https://openrouter.ai/docs/guides/best-practices/reasoning-tokens
OpenRouter and Novita use OpenAI-standard reasoning_effort values
(xhigh|high|medium|low|minimal|none), not DeepSeek-native (max|high).
The xhigh/max/highest arm previously sent 'max' for all providers,
causing HTTP 400 on OpenRouter. Split OpenRouter+Novita into their
own arm that sends 'xhigh' instead.
Account for the help modal border and padding when calculating the visible
row window, so the selected command is not clipped after scrolling.
Also wrap Up/Down navigation at the list edges and strengthen the selected
row highlight for clearer focus.
Hunter Bown