DeepSeek TUI: baixar versions anteriors

Added
- Hugging Face Inference Providers. Added huggingface as a native provider route (/provider huggingface). Supports HUGGINGFACE_API_KEY or HF_TOKEN for auth, HUGGINGFACE_BASE_URL and HUGGINGFACE_MODEL for overrides, and deepseek-ai/DeepSeek-V4-Pro / deepseek-ai/DeepSeek-V4-Flash as default models. Org-prefixed model IDs pass through.
Fixed
- Agent-mode shell error copy. The missing-tool error for shell tools now directs users to allow_shell = true instead of nudging toward YOLO mode. /config surfaces allow_shell in the Permissions section.
- Provider description. /provider command description is now neutral instead of recommending specific providers.
Community
Thanks to @xyuai for provider persistence, /logout scope clarification, provider picker key replacement, and MiMo auth cleanup work (#2714, #2715, #2717, #2718), and @RefuseOdd for configurable path_suffix support on OpenAI-compatible endpoints (#2558).

Added
- SiliconFlow China region provider. Added the siliconflow-CN provider variant for the China regional endpoint, sharing the existing [providers.siliconflow] credentials and SILICONFLOW_API_KEY slot instead of creating a second credential namespace; the provider picker and registry docs now expose the regional route explicitly (#2588, #2615).
- Multimodal /attach image forwarding. Attached images are now sent as OpenAI-compatible image_url content blocks so multimodal providers can actually see image attachments (#2584, #2587, #2607).
- Sub-agent lifecycle hooks and runtime metadata. Sub-agent spawn/complete hook events, mode-change runtime messages, mode metadata on turns, localized context-inspector strings, and drag-to-resize sidebar width are included in this release slice.
Fixed
- Sub-agents now auto-cancel after stale heartbeats. Running sub-agents track manager-visible progress and are auto-cancelled after the configurable [subagents] heartbeat_timeout_secs window (default 300s), releasing their concurrency slot and unblocking parent turns that would otherwise wait forever (#2603, #2614, #2620).
- Work panel state survives transient lock misses. The sidebar caches the last successful Work summary so checklist and strategy progress no longer disappear into "Work state updating..." while the engine briefly owns the shared todo/plan locks (#2606, #2616).
- SiliconFlow-CN no longer breaks main. Filled the missing CLI provider exhaustiveness arms and removed the duplicate/unreachable TUI config arms left by the #2615 landing; direct auth now stores the China-region variant in the shared SiliconFlow provider table (#2616, #2618, #2619).
- v0.8.51 image-attach closure corrected. The /attach multimodal fix landed after the v0.8.51 tag, so this release is the first version that actually contains it for users installing from the published release line (#2584, #2607).
- Legacy SSE MCP reconnects are retryable again. Closed or reset POST /messages requests on stale legacy SSE sessions now trigger the same reconnect-and-retry path as closed SSE streams, removing a release-gate flake and matching the intended recovery behavior (#2597).
- Cache-hit cost accounting uses one telemetry source. Mixed DeepSeek prompt_cache_hit_tokens and OpenAI-style cached_tokens usage payloads no longer infer cache misses from the wrong hit count, avoiding inflated TUI cost estimates on cached DeepSeek turns (#2567, #2609).
- Cygwin/MSYS2 config paths honor exported $HOME. CodeWhale and legacy DeepSeek config roots now prefer a non-empty $HOME before falling back to the platform home resolver, while CODEWHALE_HOME remains the strongest explicit override (#2369, #2610).
Community
Thanks to @xyuai (#2587), @IcedOranges (#2584), @BH8GCJ (#2588), @shenjackyuanjie (#2618, #2619), @idling11 (#2606, #2616), @AresNing (#2578), @caiyilian (#2567), @buko (#2369), @gordonlu, @encyc, and @simuusang (#2603, #2620) for reports, patches, retesting, and release-stabilization signals that shaped this pass.

Added
- Arcee AI as a direct provider. New [providers.arcee] config block and ARCEE_API_KEY / ARCEE_BASE_URL / ARCEE_MODEL environment variables, wired through CLI auth (codewhale auth set --provider arcee), the TUI provider picker, and the model registry. The default direct-API model is trinity-large-thinking (reasoning-capable, 262K context and 262K max output); trinity-large-preview (262K context, non-reasoning) and trinity-mini (128K context) are also selectable. OpenRouter's arcee-ai/trinity-large-thinking route remains separate.
- Arcee Cloudflare-WAF compatibility. The opening turn to the Arcee gateway uses a benign read-only tool surface (read_file, list_dir, file_search, grep_files, git_status, git_diff, checklist_write, update_plan) and splits example payloads such as python -c … out of the system prompt, so the WAF does not reject the first request; the full tool catalog stays reachable through tool-search. trinity-large-thinking's reasoning_content is recognized and replayed on tool-call turns.
- Expanded model catalog. Added context-window, max-output, and reasoning-capability metadata for additional model IDs, including qwen/qwen3.6-flash, qwen/qwen3.6-plus, qwen/qwen3.6-max-preview, and Xiaomi MiMo v2.5 chat/ASR/TTS variants; trinity-large-preview's context window was corrected to 262K.
- Provider-aware model picker. The picker groups models by provider, shows per-model hints, and remembers a saved model per provider.
Changed
- Auto-compaction is now percentage- and model-aware. The per-model threshold helper is compaction_threshold_for_model_at_percent(model, percent) (replacing the effort-based variant), and the default auto_compact_threshold_percent is 80%. Auto-compaction defaults on for models with a context window of 256K or smaller and stays opt-in for 1M-token models (e.g. DeepSeek V4) to protect prefix-cache economics, unless the user has explicitly set auto_compact.
- Clearer provider/gateway errors. HTTP error bodies are sanitized before display — HTML interstitials and Cloudflare "Access Denied" pages collapse to a one-line reason (with the ray/error ID) instead of dumping raw markup into the transcript — and 403s are split into authentication vs. authorization (gateway/WAF block) categories.
- The invalid-model error now names the active provider and lists Arcee among the options.
Removed
- The session "cycle" / checkpoint-restart system. Removed the /cycles, /cycle <n>, and /recall commands, the recall_archive tool, the cycle-handoff briefing prompt, the sidebar "cycles" lines, and the cycle_manager engine plumbing (EngineConfig.cycle, Event::CycleAdvanced, seam-manager cycle thresholds and flash briefings). Long sessions no longer auto-reset their context at a fixed token boundary — reclaim budget with /compact or model-aware auto-compaction instead. Existing on-disk cycle archives are left untouched but are no longer read or written.
Fixed
- Assistant turns no longer leave an orphaned role glyph (the stray "blue dot") when a turn streams only whitespace between reasoning and a tool call.
- Scrolling the mouse wheel over the right-hand sidebar no longer leaks into the transcript scroll.
- The sidebar hover tooltip now appears only for truncated lines, sits below the cursor, and uses a neutral surface color instead of the warning-orange highlight that overlapped neighbouring rows.
- Corrected the README's description of the Constitution (Article VII is the hierarchy itself; Article II's truth duty overrides even a user request) to match prompts/base.md.
- Repaired release-blocking unit and integration tests left failing by the cycle-removal and compaction-threshold refactors (relay instruction, model-reject message, compaction budget, mock-LLM threshold helper).
- Fixed DEC private-mode CSI fragment leakage into composer text after terminal resets, restoring clean prompt editing (#2592).
- The engine now recovers from turn-level panics instead of killing the main event loop, keeping the session ali

Added
- Added a Windows NSIS installer release artifact and classroom/lab deployment checklist, harvested from #2045 for #1987. The release workflow now builds CodeWhaleSetup.exe from the canonical Windows binaries, and the installer adds/removes only the exact current-user PATH entry.
- Added deterministic session timestamps in session listings, receipt-export boundary docs, and current-model turn metadata for routed/auto sessions.
- Added exact AtlasCloud provider-hinted model ID pass-through for explicit vendor/model-id selections, harvested from #2569 without freezing a brittle provider catalog.
- Added Xiaomi MiMo speech/TTS support with a codewhale speech CLI command, tts tool alias, and config wiring for voice-design and voice-clone models, harvested from #2560.
- Added a three-zone immutable prefix diagnostic layer (FrozenPrefix Phase 2) that logs cache-prefix drift at debug level without blocking requests, harvested from #2514.
- Added a Cache Guard CI integration test suite simulating prefix-cache behaviour across nine scenarios, gated behind CODEWHALE_CACHE_GUARD=1, harvested from #2503.
- Added a plan-mode byte-stability invariant test verifying that the tool catalog head remains byte-identical across mode toggles, harvested from #2519.
- Localized all 15 /queue command messages across 7 shipped locales, harvested from #2568.
- Added localized FanoutCounts MessageId for i18n of the aggregate worker stats line in fanout cards, harvested from #2566.
- Added contribution gate CI workflows (PR gate, issue gate, contributor approval) with a dry-run mode, harvested from #2565.
Changed
- Hardened theme repainting and sidebar color use so theme switches do not leave stale Whale-dark panel colors behind.
- Made legacy config migration visible when CodeWhale copies old DeepSeek-era config into the CodeWhale config path.
Fixed
- Fixed /context to use the effective routed model for context-window budgeting, so DeepSeek V4 routes report the 1M-token window and legacy DeepSeek routes keep the 128K fallback.
- Fixed npm wrapper version output so --version prefers the installed binary version instead of stale package metadata when both are available.
- Fixed multiline composer arrow navigation so holding Up/Down at the first or last line no longer replaces the current draft with prompt history.
- Fixed foreground exec_shell output collection so timeout and inherited-pipe cleanup cannot wedge later tool calls behind the global tool lock.
- Clarified the English DeepSeek account-balance footer chip from bal to balance so it is less likely to be mistaken for session spend.
- Fixed truncated subagent tool calls and repeated truncated subagent responses so they return model-visible errors instead of silently failing.
- Moved Paste to the first position in the right-click context menu so users copying text from the output area can paste with a single left-click instead of navigating past cell-specific actions.
Community
Thanks to @ZhulongNT (#2045), @cyq1017 (#2521, #2536, #2537, #2559, #2562, #2563, #2564), @HUQIANTAO (#2527, #2519, #2503), @lucaszhu-hue (#2569), @idling11 (#2573), @encyc (#2514), @xyuai (#2560), @gordonlu (#2568, #2566), and @nightt5879 (#2565) for the work harvested into this release pass. Thanks also to issue reporters and verification helpers including @New2Niu (#2561), @buko (#2533, #2369), @wywsoor (#2494), @ctxyao (#2556), @Dr3259 (#2380), @caiyilian (#2567), and @chinaqy110 (#2571) for reports and acceptance details that shaped these fixes, plus the WeChat/Chinese UX reports relayed during the final triage pass.

Added
- Added the missing [providers.moonshot] example block for Moonshot/Kimi, documented completion_sound, and refreshed the tool-surface docs for the current registry, including finance, web.run, git history tools, memory, OCR, and other registered tools.
Changed
- Hardened prefix-cache fingerprints to hash API-visible tool schema details, not just tool names, so schema and description drift invalidates cached prefixes before it can confuse model calls (#2264).
- Kept finance registered independently from web-search tools and prevented duplicate web/patch tool registration in agent and YOLO modes.
Fixed
- Fixed the DeepSeek V4-Pro cost estimate after the 2026-05-31 pricing cutoff: the post-promotion official rate remains one quarter of the original price, so CodeWhale no longer shows roughly 4x too much after June 1 (#2489).
- Fixed Kimi/Moonshot tool schema normalization by moving parent type fields into anyOf/oneOf items, with regression coverage for nested schema shapes that could otherwise still fail Kimi validation (#2438).
- Fixed raw ANSI/SGR fragments leaking into footer, shell-label, and sidebar activity text during active tool execution (#2481).
- Fixed [tui] config parsing when status_items is omitted, restoring the documented default footer order for older and hand-written configs (#2483).
- Fixed a shell env-scrubbing test so it does not depend on the user's default shell understanding POSIX parameter expansion.
- Removed stale qwen/qwen3.7-max references left in config.example.toml after the v0.8.48 preset removal.
Community
Thanks to @idling11 (#2480, #2485), @reidliu41 (#2493), @hongqitai (#2495), and @encyc (#2477) for the fixes and reliability work harvested into this release.
Thanks also to reporters and verification helpers whose issues shaped the release: @A-Corner (#2438), @taiwan988 (#2483), @AiurArtanis (#2489), and @Hmbown (#2481).

Added
- Recent large OpenRouter model presets. Added completions, aliases, routing metadata, and docs for Arcee Trinity Large Thinking, Qwen 3.7 Max, Xiaomi MiMo v2.5, Qwen 3.6 open-weight models, Kimi K2.6, GLM 5.1, Tencent Hy3, Gemma 4, and Nemotron (#2461).
- Provider and web-search expansion. Added Xiaomi MiMo provider support, SiliconFlow, AtlasCloud static models, Volcengine Ark search, Baidu AI Search, provider-picker coverage, and richer custom-provider docs (#2246, #1868, #2421, #2429, #2371, #2394, #2287).
- Workflow and tool ergonomics. Added the external-tool abstraction, pluggable TUI tool registry, custom slash-command allowed-tools enforcement, opt-in Unix socket hook sink, message-submit transform hooks, tool-cache introspection, and cache warmup-key tracking (#2294, #2420, #2326, #2430, #2434, #2423, #2424).
- TUI workflow features. Added /purge, /hunt, thinking fold/unfold, terminal-transparent/Solarized Light/Claude themes, footer branch display, macOS notifications, intent summaries before approval prompts, and the mobile runtime smoke/QR workflow (#2387, #2306, #2385, #2276, #2270, #2267, #2347, #2260, #2389, #2403).
- Platform and localization coverage. Added RISC-V prebuilt-binary support, Vietnamese localization, Java/Vue language-server defaults, runtime event envelopes, task migration/env isolation fixes, and state-message parent IDs for future forks (#2383, #2358, #2367, #2252, #2272, #2308).
Changed
- Release hardening. CI now runs clippy/docs checks, web frontend lint and type checks, provider-registry drift checks, broader crate docs, and a large unit-test pass across core, MCP, TUI core, app-server, and web helpers (#2443, #2444, #2274, #2446-#2460, #2440, #2441, #2450, #2448, #2454).
- Prompt, context, and model routing behavior. Stabilized project-context pack ordering, exposed the auto route in turn metadata, allowed embedders to override or inline constitutional instructions, moved volatile environment context below the prompt boundary, and used the effective model for compaction budgeting (#2418, #2410, #2356, #2311, #2314, #2437).
- Execution policy foundation. Added typed ask-rule groundwork and kept task_shell_start gated behind allow_shell, preparing the permission UI path without broadening default shell access (#2404, #2384).
Fixed
- Windows and shell reliability. Suppressed alt-screen logging on Windows, added the Windows batch launcher path, kept task shell tools eagerly loaded, loaded exec-shell companion tools consistently, covered controlling-terminal behavior, and improved shell tool availability errors (#2259, #2295, #1861, #2271, #2331, #2414, #2412).
- Session and transcript durability. Fixed hidden-worktree discovery saturation, stalled in-progress turn recovery, session persistence truncation, cached-transcript user-message highlighting, large tool-output receipting, session-detail block serialization, and deterministic composer history flushing (#2273, #2329, #2283, #2395, #2386, #2297, #2265, #2375).
- Provider and UI polish. Accepted custom model IDs in /model for non-DeepSeek providers, fixed Feishu per-chat model switching, localized context-menu labels, updated terminal tab naming, kept picker selections visible, allowed slash-space composer messages, and improved PDF text cleanup (#2280, #2149, #2320, #2319, #2324, #2316, #2266).
- Security and dependency hygiene. Bumped tar and qs, trusted fake-IP placeholder ranges only when explicitly configured, decoded Bing result URL entities, fixed legacy MCP SSE connections, and replaced manual tool error display code with thiserror derives (#2364, #2425, #2355, #2245, #2301, #2442).
Community
Thanks to contributors whose PRs landed or were harvested in this release: @HUQIANTAO (#2246, #2257, #2267, #2384, #2385, #2389, #2403, #2440-#2460), @h3c-hexin (#2311, #2313, #2314, #2354, #2355, #2356), @reidliu41 (#2291, #2316, #2324, #2357, #2366, #2386, #2431), @aboimpinto (#2290, #2294, #2295, #2326, #2433), @donglovejava (#2302, #

Added
- CODEWHALE_* env aliases. CODEWHALE_PROVIDER, CODEWHALE_MODEL, and CODEWHALE_BASE_URL are public product-scoped aliases that take precedence over the legacy DEEPSEEK_* forms. The DEEPSEEK_* names remain accepted for back-compat.
- Platform archive bundles. Release artifacts now ship as per-platform archives (tar.gz for Linux/macOS, .zip for Windows) containing both codewhale and codewhale-tui binaries plus an install script. No more downloading two loose files and guessing which ones to pick (#2193).
- Windows portable archive. codewhale-windows-x64-portable.zip ships the two binaries without an install script for USB-stick distribution (#2193).
- Web install download tile. The website install page now shows a platform-aware download tile with arch detection, SHA256 checksum display, and China mirror links, instead of burying the download behind the Cargo instructions (#2192).
- Whale dark palette refresh. Better contrast and layer separation across the TUI color scheme (#2197).
- Auto-collapse finished sub-agents. Completed sub-agent sessions now collapse automatically in the sidebar, reducing noise during long sessions (#2195).
- Shell-running status chip. A ⏳ shell running chip appears in the TUI footer while background shell tasks are active (#2194).
- Sandbox process hardening (Linux). PR_SET_DUMPABLE=0, NO_NEW_PRIVS, and RLIMIT_CORE=0 are applied at shell startup to harden child processes against inspection and privilege escalation (#2183).
- CONTRIBUTING.md cross-links. Issue and PR templates are now cross-linked from CONTRIBUTING.md to improve contributor onboarding (#2203).
Changed
- DeepSeek-first focus. v0.8.46 refocuses on delivering the highest-quality experience on DeepSeek first. Additional first-class provider paths are planned for v0.9.0 after the core DeepSeek workflow is solid.
Fixed
- Model name casing preserved. normalize_model_name_for_provider no longer lowercases user-set model names such as DeepSeek-V4-Flash, preventing API lookup failures on case-sensitive backends (#2109).
- Esc in model picker applies selection. Dismissing the model picker with Esc now applies the last-highlighted choice instead of reverting (#2196).
- Web install downloads both binaries. The install-binary.tsx snippet now fetches both codewhale and codewhale-tui, fixing the MISSING_COMPANION_BINARY trap on fresh npm installs (#2191).
- grep_files skips large directories. The pure-Rust search tool now skips known-large directories (.git, node_modules, target) before walking, preventing hangs on deep or slow filesystems.
- Version-update hint uses semver. The update notification in the footer now compares versions semantically instead of lexicographically, so 0.8.10 > 0.8.9 is recognized correctly.
- CVE-2026-8723 in feishu-bridge. Bumped qs to >=6.15.2 in the Feishu bridge integration (#2198).
Community
Thanks to new contributors whose PRs landed in this release: @donglovejava (#2154, #2163, #2166, #2167, #2168), @encyc (#2152), @saieswar237 (#2178), @sximelon (#2174), @nanookclaw (#2135), @Sskift (#2119), @xin1104 (#2105), @mrluanma (#2059), @Lellansin (#2055), @zhuangbiaowei (#2145), @aboimpinto (#1872), and continuing contributors @reidliu41, @cyq1017, @idling11, @h3c-hexin, @wdw8276, @zlh124, and @jeoor.

Added
- RLM session objects. rlm_open can now load session:// refs, exposing the active prompt, history, and session data as symbolic objects inside RLM REPLs (#2047).
- Command palette voice input. The command palette can launch a configured speech-to-text helper and show footer status while transcription runs (#2047).
- Moonshot/Kimi provider. Moonshot/Kimi is now a first-class provider, including API-key auth, model completion, CLI auth, secret-store integration, and optional Kimi CLI credential reuse.
- Deterministic whale-species sub-agent names. Sub-agents now get stable, human-readable whale-species nicknames (e.g. "Beluga", "Orca") while preserving the raw agent ID in the popup (#2035, #2016).
- /balance command scaffold. Registered the /balance slash command as a placeholder for future provider billing queries (#2035, #2019).
- Readable /restore snapshot labels. Snapshot labels now include the originating user prompt so restore listings are easier to identify. Thanks @idling11 (#2111).
- Sidebar hover tooltips. Truncated Work and Tasks sidebar lines now expose their full text on hover. Thanks @idling11 (#2110).
Changed
- AGENTS.md is now maintainer-local. The project instructions file no longer ships as a tracked repo file; it lives in maintainer-local ignored state (#2047).
Fixed
- Sub-agent completion handoff compatibility. Completion handoffs now use a chat-template-safe role and emit before terminal updates, fixing strict OpenAI-compatible/self-hosted backends and preserving transcript ordering. Thanks @h3c-hexin and @cyq1017 (#2057, #2120).
- Self-hosted context budgeting. Sub-500K self-hosted model windows now keep a usable input budget instead of disabling preflight compaction after output reservation underflow. Thanks @h3c-hexin (#2060).
- Goal prompts start actionable. Goal-start prompts now open in an actionable state instead of requiring an extra nudge. Thanks @cyq1017 (#2097).
- Composer session title display. The composer chrome shows the current session title again and avoids grayscale luma overflow in debug builds. Thanks @wdw8276 (#2108).
- Approval prompts use a one-step confirmation flow. Enter now commits the selected approval option directly, destructive warnings remain visible, and abort cancels the active turn instead of only denying the current tool call. Thanks @reidliu41 (#2143).
- Model picker selection survives Esc. Dismissing the model picker with Esc no longer loses the highlighted selection. Thanks @reidliu41 (#2056).
- Moonshot/Kimi sessions launch from the dispatcher. The codewhale wrapper now includes Moonshot/Kimi in the TUI provider allowlist, so codewhale --provider moonshot --model kimi-k2.6 reaches the TUI instead of stopping after config resolution.
- Slash recovery no longer restores command tails in the composer. Resuming a session or recovering from a crash no longer leaves stale slash-command text (e.g. /sessions) in the composer input (#2047, #2032).
- Remembered tool approvals now update the live active turn. When the "remember" checkbox is set on an approval dialog, the active turn's auto-approve flag flips immediately instead of waiting for the next turn. Thanks @gaord (#2047, #2041).
- YAML block scalars in SKILL.md frontmatter. Multi-line descriptions using > or | indicators are now parsed correctly — folded block scalars join non-empty lines with spaces, literal scalars preserve newlines, and all three chomping modes (strip/clip/keep) are supported. Thanks @zlh124 (#1908, #1907).
- User messages highlighted in the transcript. User-authored messages now render with a full-row background in the live TUI transcript, making it easier to scan prior turns. Assistant and system messages are unaffected. Thanks @reidliu41 (#1995, #1672).
- Cancellable list_dir and file_search. Long directory walks and file searches now respond to user cancel/stop requests with a 30-second fallback timeout, preventing the TUI from hanging on deep or slow filesystems (#2035).

Added
- codew convenience alias. codew is a short-form command that silently forwards to codewhale. Six fewer keystrokes, same binary. Ships with the Rust codewhale-cli crate and the npm codewhale package (#2013).
- Session picker inline rename. Press r in the session picker (Ctrl+R) to rename the selected session inline. Type the new title, Enter to confirm, Esc to cancel (#1600).
- Plan detail display. The "Plan Confirmation" modal now shows the plan explanation and step list from update_plan so you can review what was proposed before accepting (#834).
- Agent team UX. Delegate cards in the transcript now show human-readable roles (scout, builder, reviewer, verifier, executor) and the completion summary instead of raw agent_xxx IDs (#1981).
- --continue / -c CLI flag. codewhale --continue resumes your most recent interactive session for the current workspace.
Changed
- App state migrates to ~/.codewhale/. New installs write product-owned state (config, sessions, tasks, skills, logs, etc.) under ~/.codewhale/. ~/.deepseek/ continues to work as a compatibility fallback — no data loss, no forced migration. CODEWHALE_HOME and CODEWHALE_CONFIG_PATH env vars are now supported alongside existing DEEPSEEK_* vars (#2011).
- Project config overlay prefers .codewhale/config.toml before .deepseek/config.toml. Both are read; the CodeWhale root takes precedence.
- Doctor reports active state root and whether legacy ~/.deepseek/ state is also present.
- README contributor acknowledgements are current for this release. Thanks @jeoor, @LING71671, and @ousamabenyounes for the fixes and reports now reflected in the public credits.
- Harvested-contribution credit audit completed. The README Thanks list now includes previously missed community helpers whose code, reports, or review notes were already credited in older changelog entries but not in the public contributor surface: @mvanhorn, @krisclarkdev, @tdccccc, @LittleBlacky, @AnaheimEX, @THatch26, @alvin1, @knqiufan, @IIzzaya, @duanchao-lab, @imkingjh999, @eng2007, @chennest, @kunpeng-ai-lab, @asdfg314284230, @maker316, @lalala-233, @muyuliyan, @czf0718, @MeAiRobot, @tiger-dog, @MMMarcinho, @lucaszhu-hue, @sandofree, @zhuangbiaowei, @NorethSea, @Jianfengwu2024, @Fire-dtx, @oooyuy92, @qinxianyuzou, @tyouter, @xulongzhe, @YaYII, @47Cid, and @JafarAkhondali.
- Harvest guidance now requires GitHub-visible attribution. Maintainer harvests should preserve the original commit author where possible or add Co-authored-by trailers from the original PR commits, in addition to the existing Harvested from PR #N by @handle trailer and changelog credit.
- Enter now steers when busy-waiting. When the model is busy but not actively streaming (waiting on tool results, sub-agents, or shell commands), pressing Enter tries to steer your message into the current turn instead of silently queueing it. During active streaming, Enter still queues to avoid interrupting in-flight reasoning (#2009).
Fixed
- /save no longer creates repo-local session_*.json. Default saves now go to the managed sessions directory instead of the current workspace. Explicit /save path/to/file.json exports still work as before (#2010).
- Boot-time session prune caps managed sessions at 50 on every startup, preventing unbounded growth of ~/.codewhale/sessions/.
- Checkpoint path resolution no longer hardcodes ~/.deepseek/ — uses the resolved session directory instead.
- Plain startup no longer auto-opens the session picker. codewhale and codew start in a fresh composer again even when saved sessions exist. Use /sessions, Ctrl+R, --resume, or --continue when you want to resume.
- Work sidebar now refreshes immediately after checklist_write, checklist_update, and update_plan tool calls, matching the existing todo_write behavior instead of relying on the 2.5s periodic poll (#1787).

Fixed
- grep_files now respects the cancellation token. Long-running file searches cancel promptly instead of running to completion after the user aborts (#1839). Thanks @LING71671.
- npm installer stream-pause race condition fixed. The install script now pauses HTTP response streams immediately, preventing early data loss that caused "Invalid checksum manifest line" errors (#1860). Thanks @jeoor.
- Ctrl+Z restores the last cleared composer draft. Pressing Ctrl+Z in an empty composer recovers the text that was last cleared with Ctrl+U or Ctrl+S, matching the muscle memory users expect from other editors (#1911). Thanks @LING71671.
- Clipboard works on non-wlroots Wayland compositors. The Linux clipboard path now tries wl-copy before arboard, fixing silent copy failures on niri, River, cosmic-comp, and GNOME mutter (#1938). Thanks @ousamabenyounes.
Added
- Goal mode ships as a persistent objective surface. Orthogonal to Plan / Agent / YOLO execution modes. Use /goal <objective> to set a goal, /goal done to mark it complete. Goal status appears in the Work sidebar with elapsed time. Alt+G toggles Goal mode; /mode goal or /mode 4 activates it from the command line (#1976).
- Post-turn receipts cite evidence for every completed turn. When a turn finishes, a receipt line shows in the transcript tail with a summary of tool calls, file changes, and evidence that supports the agent's claims. Tool evidence is collected per-turn and flushed on new dispatch.
- Stall reason classification. When a turn has been running for more than 30 seconds, the footer now appends a classified reason: "waiting for model", "tools executing", "sub-agents working", "compacting context", or "waiting — no recent activity".
- Decision card widget for structured user input. When Brother Whale needs a choice, it surfaces a bordered card with numbered options, keyboard navigation (1-9 / j/k / arrows), and Enter/Esc to confirm or cancel.
- Tasks sidebar now shows fuller turn IDs and supports copy-to-clipboard. Turn ID prefixes are widened from 12 to 16 characters for disambiguation, background job status is presented as "X running, Y completed" instead of ambiguous "X active (Y running)", and y / Y yank affordances copy the current turn ID or full status line to the system clipboard (#1975).
Changed
- Contributor count and acknowledgement surfaces refreshed. The website fallback contributor count now reflects 98 live GitHub contributors (up from the stale 91). All three README translations (English, 中文，日本語) now include 30+ previously unlisted contributors whose PRs were merged since April 2026.
- README and web surface rebrand refinements. Crate descriptions, npm package text, and website copy now consistently position CodeWhale as open-model-first and provider-spanning, with DeepSeek V4 as the first-class path.
- New contributor names added to README acknowledgements. Thanks to @Apeiron0w0, @aqilaziz, @ChaceLyee2101, @ComeFromTheMars, @CrepuscularIRIS, @dst1213, @eltociear, @fuleinist, @greyfreedom, @h3c-hexin, @heloanc, @hxy91819, @J3y0r, @JiarenWang, @jinpengxuan, @KhalidAlnujaidi, @laoye2020, @lbcheng888, @linzhiqin2003, @Liu-Vince, @lixiasky-back, @pengyou200902, @punkcanyang, @Rene-Kuhm, @SamhandsomeLee, @sockerch, @sternelee, @Wenjunyun123, @whtis, and @wuwuzhijing for the translations, typo fixes, docs polish, and small UX improvements that landed across the 0.8.42 → 0.8.43 cycle.
Security
- Thinking blocks can be collapsed/expanded via keyboard. Space on an empty composer toggles the focused thinking cell between collapsed and expanded, complementing the existing mouse right-click context menu (#1972).
- Sub-agent completion events no longer delayed to the next turn. The turn loop now drains late-arriving sub-agent completions at the final checkpoint before breaking, so child-agent sentinels surface immediately instead of appearing in the following turn (#1961).
- codewhale doctor now referenced correctly in SSE timeout errors. The error message shown wh

Changed
- CodeWhale now ships with the Brother Whale agent identity prompt. The built-in system prompt frames the agent as trusted, calm, careful, and responsible, and adds the coordination principle that great intelligence creates spaces where future intelligences can work together.
- CodeWhale positioning is clarified as DeepSeek-first and open-model oriented. README, rebrand notes, crate metadata, and npm package text now describe CodeWhale as an agentic terminal for open source and open-weight coding models while preserving the official DeepSeek provider as first-class.
- Model auto-routing is documented separately from TUI modes. README and modes docs now reserve "mode" for Plan / Agent / YOLO, describe --model auto as model/thinking routing, and name the fast deepseek-v4-flash thinking-off seam as Fin.
- Rebrand shim docs now match the v0.8.x transition window. The npm and migration notes no longer imply the legacy deepseek-tui package/shims expired immediately after v0.8.41.
Fixed
- User-authored messages render as literal plain text. Leading whitespace, whitespace-only lines, repeated spaces, and Markdown-looking # / - text now survive in transcript history, while assistant messages still render Markdown normally.
- English turns stay English after localized context. The Brother Whale identity and base language rules no longer inject native-script examples into the English prompt path, and the prompt now calls out localized READMEs, issue text, file contents, and tool results as data rather than language signals.
- Stream decode failures no longer leave the turn visually stuck. The UI now marks an active turn failed and flushes live cells as soon as the engine emits a stream error, so the sidebar/footer recover without requiring Ctrl+C (#1960).
- RLM contexts now expose _ctx. Persistent RLM REPLs bind _ctx as a compatibility alias for the loaded source alongside _context and content, and the prompt/docs call out the exact names (#1962).
- handle_read is easier to recover from. The tool keeps accepting full var_handle objects directly, adds introspect: true for size/projection hints, and validation failures now include copy-pasteable examples (#1963).
- The help picker keeps the selected row visible while scrolling. /help now budgets against the real modal body height, wraps Up/Down navigation, and uses a stronger selected-row highlight (#1964).
- Unicode git_status paths stay readable. Chinese and other non-ASCII repository paths now survive status parsing and display cleanly (#1936, #1953).
- Project-local and configured skills appear in the slash menu. Workspace skills and configured skill directories now feed the command picker instead of only the bundled set (#1955, #1956).
- Repeated Tab mode switching no longer stacks composer-obscuring toasts. The mode-switch notification now deduplicates instead of accumulating rows over the composer (#1926, #1957).
- Local tool UX surfaces are clearer. github_close_pr now has the same guarded closure workflow as issue close, handle_read redirects artifact refs to retrieve_tool_result, Plan handoffs use plainer wording, and shell rows/sidebar tasks show the actual running command instead of placeholder labels.
Thanks
Thanks to cyq (@cyq1017) for the Unicode git_status, local/configured skill discovery, and mode-switch toast fixes in #1953, #1956, and #1957. Thanks to Reid (@reidliu41) for the help picker scrolling and selection fix in #1964.

Changed
- Project renamed to codewhale. The canonical CLI dispatcher is now codewhale (was deepseek) and the TUI runtime is codewhale-tui (was deepseek-tui). The 14 workspace crates are renamed from deepseek-* / deepseek-tui-* to codewhale-* / codewhale-tui-*. The npm wrapper package is now codewhale (was deepseek-tui). See docs/REBRAND.md for migration notes.
- DeepSeek provider integration is unchanged. DEEPSEEK_* env vars, model IDs (deepseek-v4-pro, deepseek-v4-flash, the legacy deepseek-chat / deepseek-reasoner aliases), the https://api.deepseek.com host, and the ~/.deepseek/ config directory are all preserved.
Deprecated
- The deepseek and deepseek-tui binary names continue to ship as tiny shims that print a one-line warning and forward argv to the renamed binaries. They will be removed in v0.9.0.
- The deepseek-tui npm package continues to publish for one release cycle as a no-bin deprecation shim whose postinstall directs users to npm install -g codewhale. It will be removed in v0.9.0.
Fixed
- Windows CI spillover tests are isolated. Tool-result deduplication tests now use a temporary spillover root guarded by the existing global spillover mutex, removing the shared-state race that made Windows CI fail unrelated PRs (#1943).
- Terminated sub-agents keep agent_eval recoverable. Evaluating a completed child session now returns the available transcript result instead of losing the final output (#1738, #1928).
- Bare @/ completions no longer freeze the TUI. File-mention completion skips bare separator and dot tokens so Windows/WSL2 workspaces do not trigger an eager 4096-entry filesystem walk on the UI thread (#1921, #1929).
- Enter paths avoid synchronous UI-thread waits. Composer history writes, offline queue persistence, feedback URL launching, and clipboard fallback helpers now run off the hot Enter path where appropriate (#1927, #1931, #1940, #1941, #1944).
- tmux and screen sessions stop idling as terminal activity. Terminal multiplexers now force low-motion behavior and pin the fallback footer label so passive animations do not trip activity monitors (#1925, #1942).
- Composer sanitization catches OSC 8 and Kitty fragments. The input sanitizer now strips common hyperlink and keyboard-protocol fragments that leaked into drafts while preserving ordinary prose (#1915, #1933).
- The Work sidebar hides stale completed tasks. Terminal task records older than the current session and outside the recent-completion window no longer crowd active Work sidebar rows (#1913, #1930).
- V4 Pro pricing docs reflect permanent rates. The English, Simplified Chinese, and Japanese READMEs now describe the V4 Pro pricing change as permanent instead of temporary (#1923, #1932).
Thanks
Thanks to OpenWarp (@zerx-lab) for prioritizing codewhale support and collaborating on terminal-agent UX. Thanks to @leo119 for the update-command documentation lineage now preserved through the rename.

Added
- Configurable sub-agent per-step API timeout. A new [subagents] api_timeout_secs setting in ~/.deepseek/config.toml controls how long each sub-agent step will wait on a DeepSeek create_message response before falling back. The value is clamped to 1..=1800; 0 or unset preserves the legacy 120-second default, so existing installs see no behavior change. Long-thinking children (e.g. heavy plan or review work behind agent_open) can extend the timeout without recompiling (#1806, #1808).
- Delegated file-write permissions for write-capable sub-agent roles. implementer and custom sub-agents may now run Suggest-level write tools (write_file, edit_file, apply_patch) without the parent runtime being auto-approved. Read-only stances (explore, plan, review, verifier) and the default general role still bounce approval-gated tools so they can't quietly mutate the workspace, and Required-level tools (shell, etc.) still need parent auto-approve regardless of role. Pick implementer (or pass an explicit custom allowlist) when the delegated task needs to land file changes (#1828, #1833).
- Experimental Fin fast-lane tool agents. tool_agent opens a durable child session on DeepSeek V4 Flash with thinking forced off for simple tool-bound work such as OCR, file/search lookups, fetches, and command probes. It uses the existing agent_eval / agent_close lifecycle and mailbox token-usage stream, so sub-agent cost accounting stays on the same path as normal agent_open sessions.
Fixed
- WSL2 and headless Linux startup no longer blocks on clipboard init. The TUI now defers clipboard initialization so machines without an X server can reach the first frame instead of hanging on a blank screen (#1773, #1772).
- Windows alt-screen output stays clean when RUST_LOG is set. Runtime tracing is routed away from the interactive buffer so logs no longer leak into the TUI display (#1774, #1776).
- OpenAI-compatible custom model names are preserved. Non-DeepSeek providers now pass explicit model names through instead of rewriting them to a DeepSeek default (#1714, #1740).
- Wanjie Ark is a first-class provider. --provider wanjie-ark, the TUI provider picker, deepseek auth, doctor, and config files now target Wanjie's OpenAI-compatible MaaS endpoint with pass-through model IDs and Wanjie-specific env vars.
- DeepSeek reasoning replay works through OpenAI-compatible endpoints. DeepSeek models selected under the generic openai provider now replay prior reasoning_content consistently and classify streamed reasoning the same way the replay path does (#1694, #1739, #1743).
- Thinking-only turns no longer disappear. If a clean turn ends with thinking but no final answer text, the UI now surfaces a clear status instead of silently ending the turn (#1727, #1742).
- Windows cmd /C preserves quoted shell arguments. Commands such as git commit -m "feat: complete sub-pages" now round-trip through the Windows shell wrapper without losing the quoted message (#1691, #1744).
- Home/End are line-local inside multiline composer drafts. The keys now jump to the current input line boundary before falling back to transcript navigation (#1748, #1749).
- Ctrl+C restores the canceled prompt reliably. Canceling a streaming turn puts the submitted prompt back in the composer and suppresses late stream events from drawing stale output (#1757, #1764).
- Compaction recovers from cache-aligned summary context overflow. When a cache-preserving summary request itself exceeds the provider context window, compaction retries with the bounded formatted summary path instead of failing with a 400 "compression command failed" style error.
- Terminal sub-agent sessions expose full transcript handles. Completed and canceled child agents now store the full child message transcript behind transcript_handle, so the parent can inspect details with handle_read instead of relying only on a lossy summary (#1738).
- Forked saved sessions now keep visible lineage. deepseek fork records the parent session id and

Fixed
- Feishu/Lark bridge startup order is guarded. The bridge now keeps ThreadStore initialized before startup opens persisted thread state, with a regression test to prevent moving it below its first use.
- /model picker opens instantly with the curated list again. Reverted the v0.8.38 live-catalog rework: the picker no longer makes a blocking network call on open and once again shows the curated auto / deepseek-v4-pro / deepseek-v4-flash rows. The /models command still lists the live provider catalog.
- "Approve for session" groups by command family again. Session approvals are keyed by a lossy, arity-aware fingerprint once more, so approving cargo build also covers cargo build --release. Denials keep the exact per-call fingerprint from #1617, so denying one call no longer over-blocks later, different calls to the same tool.
- Docker first-run state directories are writable. The image now pre-creates /home/deepseek/.deepseek with deepseek ownership so the documented named-volume launch can create runtime thread state on first use (#1684).
- Runtime API system prompt overrides survive the first turn. Threads created with a system_prompt override now keep that prompt through mode/context refreshes before the model request is built (#1688).
- Compaction keeps a user text query in tool-heavy histories. Automatic compaction now pins the latest user text message when the retained tail only contains tool calls/results, avoiding OpenAI-compatible Jinja template failures on the next request (#1704).
- Pager jumps land at the visible bottom. Pressing G or End in the pager no longer overshoots the render clamp, so k/Up scrolls upward immediately afterward, and mouse wheels now scroll pager overlays directly (#1706, #1716).
- Mouse-wheel-as-arrow scrolling preserves composer drafts. When composer_arrows_scroll is enabled, Up/Down now scroll the transcript even with text in the composer instead of replacing the draft with input history (#1677).
- Multiline composer arrows move between input lines. Plain Up/Down now move the cursor within multiline drafts before falling back to input history, while single-line mouse-wheel-as-arrow scrolling remains unchanged (#1721).
- Third-party reasoning_content streams no longer corrupt text output. Generic OpenAI-compatible providers that stream answer text in reasoning_content now render it as normal text unless the selected provider is one whose reasoning-content semantics are supported (#1673).
- macOS system theme detection recognizes Light mode. When COLORFGBG is missing or unusable, theme = "system" now falls back to macOS appearance detection and treats a missing AppleInterfaceStyle key as Light mode (#1670).
- rlm_open accepts schema-filled blank source fields. Empty file_path, content, and url strings now count as absent, so calls that provide one real source no longer fail the exactly-one-source validator (#1712).
- Resize keeps transcript paging usable immediately. After a terminal resize, PageUp/PageDown now use the resized viewport height instead of falling back to one-line jumps before the next render (#1724).
- ACP responses stringify JSON-RPC ids. serve --acp now returns string ids even when clients send numeric ids, matching Zed's stricter ACP client expectations (#1696).
Thanks
Thanks to Matt Van Horn (@mvanhorn) for the Docker first-run permission fix in #1699 and the runtime system-prompt regression tests harvested from #1702. Thanks to Kristopher Clark (@krisclarkdev) for the compaction user-query preservation fix in #1704. Thanks to Stephen Xu (@wlon) for the pager jump-bottom fix in #1706. Thanks to tdccccc (@tdccccc) for the composer scroll fix in #1715 and pager mouse-wheel support in #1716. Thanks to Paulo Aboim Pinto (@aboimpinto) for the multiline composer arrow navigation tests harvested from #1719. Thanks to LittleBlacky (@LittleBlacky) for the provider-gated reasoning_content stream fix in #1680. Thanks to Eosin Ai (@Aitensa) for the macOS system appearance fallback i

A bugfix + refactor release picking up the backlog after the v0.8.23 security release. Big thanks to wplll (cache-aware prompt + /cache inspect), Liu-Vince (MCP pagination diagnosis), @Giggitycountless (snapshot cap proposal), and to issue reporters @SamhandsomeLee, @barjatiyasaurabh, @tyculw, @hongyuatcufe, and @ljlbit for the bugs fixed below.
Fixed
- Mouse-wheel scroll survives focus toggles — on macOS, switching away (Cmd+Tab, opening the screenshot tool, etc.) and back can drop the terminal's mouse-tracking mode, leaving wheel scroll dead until restart. The TUI now re-arms EnableMouseCapture on FocusGained alongside the existing keyboard-mode recapture, so wheel events keep flowing after a focus round-trip.
- Workspace-local slash commands are now loaded (#1259) — user command files placed in <workspace>/.deepseek/commands/, <workspace>/.claude/commands/, and <workspace>/.cursor/commands/ are now discovered alongside the existing global ~/.deepseek/commands/. Workspace-local commands shadow global by name, matching the precedence model already used for skills. Reported by @SamhandsomeLee.
- @-mention completion finds AI-tool dot-directories — files inside .deepseek/, .cursor/, .claude/, and .agents/ are now discoverable in @-mention Tab-completion even when those directories are excluded by .gitignore. The fix also applies to the Ctrl+P file picker and fuzzy file resolution.
- MCP paginated discovery (#1250, #1256) — tools, resources, resource templates, and prompts from MCP servers that paginate their responses (e.g., gbrain at 5 items per page) are now fully discovered by following the MCP spec's nextCursor across all pages. Reported by @hongyuatcufe; thanks to Liu-Vince for the diagnosis and PR #1256 with the same fix shape.
- Snapshot storage has a disk-space cap (#1112) — the snapshot side repo now enforces a 500 MB hard limit. When the limit is exceeded at snapshot time, the oldest snapshots are pruned aggressively to stay under a 400 MB target. Guards against the reported 1.2 TB snapshot blowup during high-churn sessions. Reported by @tyculw; thanks to @Giggitycountless for the PR #1131 proposal that informed the hard-cap approach.
- /clear now resets the Todos sidebar (#1258) — previously /clear only reset the Plan panel; the Todos checklist persisted across clears until app restart. The fix ensures clear_todos() clears the SharedTodoList inner state. Reported by @barjatiyasaurabh.
Added
- Cache-aware prompt diagnostics + payload optimization (#1196) — adds a PromptBuilder that classifies the system prompt into static / history / dynamic layers for cache-prefix stability, plus:
- /cache inspect — shows SHA-256 hashes per layer, base static prefix hash vs full request prefix hash, static-prefix stability across turns, and first-divergence tracking. Does not print prompt text.
- /cache warmup — prefetches the stable prefix to seed the DeepSeek context cache.
- Project Context Pack injected into the stable prefix by default — a structured workspace summary (directory listing up to 4 levels / 400 entries, README excerpt up to 4 KB, config + key source file lists). Adds ~1–10 KB to every prompt depending on repo size, in exchange for a much more cacheable prefix. Default ON; disable with [context] project_pack = false in ~/.deepseek/config.toml if you'd rather keep prompts minimal.
- Wire-payload optimization: large tool outputs are budgeted, repeated identical tool outputs and <turn_meta> blocks are deduplicated with stable refs (wire-only — local session messages stay intact).
- Footer cache-hit % chip from prompt_cache_hit_tokens / prompt_cache_miss_tokens in the API response.
Thanks wplll for the design and implementation.
Changed
- Language directive strengthened against project-context bias (#1118) — the system prompt now explicitly instructs the model that project context (AGENTS.md, auto-generated instructions, file trees) is NOT a language signal. Chinese filenames in a repo no longer bias the model toward Chi