Skip to content

docs(tui): add missing changelog entries for 0.5.0-0.6.5 #511

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
hetaoBackend merged 1 commit into from
Oct 20, 2026
Merged

docs(tui): add missing changelog entries for 0.5.0-0.6.5 #511

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
358 changes: 356 additions & 2 deletions packages/tui/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,18 +7,372 @@ This changelog covers important changes that directly affect MCode TUI users. It

Internal refactors, test-only changes, and changes limited to other product surfaces are excluded.

## 0.5.5 · 2026-09-25
## 0.6.5 · 2026-10-09

### Fixed

- The output-speed readout no longer spikes at the start of a turn. Output buffered upstream and
released in one burst is now excluded, so a burst arriving within the first tenth of the request
no longer reports a rate far above what the model actually produced.

## 0.6.4 · 2026-10-08

### Fixed

- While you are scrolled back through a long conversation, output that changes or removes text
already in terminal scrollback no longer moves the viewport to the top of the session. Without
recent input, the screen repaints in place and the full history rebuild waits for your next
keystroke; readers following the latest output stay at the bottom.
- Settled one-time feedback, such as the previous run's duration note, finished `!` shell blocks,
and earlier run errors, now stays in the transcript once later output follows it instead of being
removed from the middle of the conversation, so later rows no longer shift.
- `cron once --after` accepts `w` (weeks), alone or combined with other units such as `1w2d`.

## 0.6.3 · 2026-10-08

### Added

- Customize the main Agent's system prompt at startup with `--system-prompt` or
`--system-prompt-file`, or extend it with `--append-system-prompt` or
`--append-system-prompt-file`. Available in the interactive TUI, `exec`, `exec review`, and ACP.
- Replacement and appended prompts can be combined. They apply only to the current process, preserve
runtime rules and project instructions, and do not affect Task child Agents. A missing or empty
prompt file stops the launch instead of running without it.
- Start a new Session in lightweight conversation mode with `--mode lightweight`. It sends a small
conversational system prompt and no tool schemas, and omits workspace instructions, Skills, memory
blocks, and MCP schemas, so it cannot inspect or change local files. The mode is fixed when a root
Session is created and cannot be combined with `--session` or `--continue`; child Agents,
`/compact`, and ACP stay on the standard context. Reopening a lightweight Session keeps its mode
and shows `Lightweight` in the status line.
- Each model request attempt now has a first-response bound and a stream idle bound, both 300
seconds by default. A request that passes a bound is aborted and reported as a retryable timeout,
so the normal model-request retry runs instead of the turn waiting on a provider that never
answers. Override the bounds with `MCODE_LLM_FIRST_EVENT_TIMEOUT_MS` and
`MCODE_LLM_STREAM_IDLE_TIMEOUT_MS`, where `0` disables a bound.

### Improved

- A single model request carries at most 20 images by default. Older images in the request are
replaced with a text note while the stored history keeps every image, and a provider that allows
fewer images per request can set `max_images_per_request` on the model (`api.mistral.ai` defaults
to 8). When compaction is rejected for too many images, it retries with attachments replaced by
text.
- Deterministic provider rejections, such as an image limit or an HTTP 400 `invalid_request_error`,
are now reported immediately with the upstream reason instead of being retried, and suggest
`/compact` or `/new` instead of `/retry`.

### Fixed

- While you are scrolled back, output that shrinks the layout, such as a task list collapsing when a
reply ends, no longer moves the viewport to the top of the session. The released rows are padded
and the history rebuild waits for your next keystroke.
- After the agent asks you a question, the turn now waits for your answer instead of ending as a
retryable empty-response failure.
- `/retry` during a running response now says to stop the running turn first, instead of reporting
that there is no failed response to retry.
- Clearing or pausing a Goal while a response is still running now says that the current response
keeps running and to press Esc to interrupt it, instead of implying the turn already stopped.

## 0.6.2 · 2026-10-02

### Fixed

- In a `/btw` side session, pressing Esc on an empty Composer no longer suggests editing the latest
message; `/edit` is unavailable there.
- Switching between a side session and the main conversation no longer resets the main run's
Running/Loading timer to zero.

## 0.6.1 · 2026-10-01

### Added

- `/doctor` and `/feedback` now work inside a `/btw` side session, so you can diagnose or report a
failed side response without leaving it.

## 0.6.0 · 2026-10-01

### Added

- `/retry` now works inside a `/btw` side session. When a side-session model request fails and can
be retried, run `/retry` there to resend the side session's last message without switching back to
the main conversation.

### Improved

- The official default model is now `MiniMax-M3.1-Flash-Preview` when no model has been selected.
- When a Goal is marked complete, the agent now ends with one final reply that summarizes what was
done and where the deliverables are, instead of stopping right after the status update.
- Safety refusals from the model provider are no longer retried, including on BYOK. The original
error is shown and identified as a refusal from the model provider.
- TLS record verification errors during a model request now trigger the automatic retry like other
transient network failures.

### Fixed

- In regular mode, the screen no longer jumps back to the opening logo and replays the whole
conversation while a run is in progress. Rows that can still change stay on screen above the
footer until they are final.
- The `tok/s` speed is no longer inflated when the provider hides thinking text; generation time now
starts when thinking starts.
- Fixed repeated failures after context compaction during an automatic continuation.
- Context compaction through OpenAI-compatible Chat Completions providers no longer sends an empty
`tools` list, which some backends rejected with HTTP 400.

## 0.5.10 · 2026-09-30

### Fixed

- Your own messages now appear exactly as typed in the conversation, pending instructions, and the
transcript viewer. Text such as `__pycache__`, `*`, and a leading `>` is no longer reformatted as
Markdown, and trailing blank lines no longer add empty rows.
- The TUI no longer stays stuck showing a run as in progress after Runtime has already finished it.
Enter now sends a new message instead of steering the finished run, and queued messages are no
longer held back.

## 0.5.9 · 2026-09-30

### Added

- The build-mode status line now reports `background=N`, the number of active background tasks in
the current root session. Bash still owned by its foreground tool call is not counted, and the
count holds at `1` after a turn settles until a fresh task list arrives, so a command that has
just moved to the background is not reported as `0`.
- `worktreeRefreshBeforeCreate` in `config.yaml`, on by default, fetches the selected upstream
before a new worktree is created.
- The OpenAI Codex OAuth entry in model settings is enabled by default; set `beta.codexOAuth: false`
to hide it.

### Improved

- BYOK requests now apply byte limits to large media and accumulated history context, reducing
failures caused by oversized requests.
- Long conversations load their file list and history navigation with less data, making browsing
them faster.
- The MiniMax API-key route now uses the same official model catalog as the signed-in route.

### Fixed

- Stopping a conversation with Esc also cancels the background Bash commands and subagents it owns,
and their completion notices no longer restart the conversation. Their results stay available to
the next turn. Switching sessions and `/clear` continue to leave background work running.
- Editing and rewinding a message now works after an interrupt that left only a background reminder
behind.
- Terminal errors from a failed run are cleared when the next run starts, so a successful retry no
longer shows the earlier error again.
- The thinking toggle for M3 models is sent as a thinking setting instead of a generic reasoning
effort.

## 0.5.8 · 2026-09-28

### Added

- Added M3.1-Flash-Preview to the built-in fallback model catalog.

### Improved

- Plugin Hook messages appear in the active conversation and restored history, making Hook results
visible without adding them to model context.

### Fixed

- Restoring a session with `--session` or `--continue` clears terminal scrollback left by the
previous process before replaying history, avoiding duplicate content. Normal startup preserves
shell history.
- Improved output queuing on slow terminals and waits for pending output before exiting, suspending,
or opening an external editor.

## 0.5.6 · 2026-09-28

### Added

- Custom provider setup can import models from `/models` with the API key entered during setup.
Search and select a model, then save the imported models after a successful connection test; retry
discovery or enter a model ID manually when needed.

### Fixed

- Token speed now uses actual provider output tokens and the generation window, excluding the wait
for the first token and tool execution. Streaming text estimates are no longer shown; confirmed
speed remains visible while waiting for subsequent requests or tools.
- History reloads preserve generation timing. Older messages without timing are excluded from speed
statistics, and speeds below 10 token/s retain one decimal place.
- Bash tool descriptions are displayed as plain text instead of receiving shell command syntax
highlighting; actual commands retain syntax highlighting.

## 0.5.5 · 2026-09-27

### Added

- Terminal titles show the current state, session name, and MCode, for example
`Needs approval | Fix login | MCode`, and follow session renames and switches. Notifications for
completed turns, failures, permission requests, and questions identify the session and suppress
duplicates; completion waits for the session's queue to finish. Configure them with
`tui.terminalTitle` and `tui.notifications` in `config.yaml`.

### Improved

- Bash commands that automatically move to the background now have a one-hour default and maximum
total timeout. Explicit background tasks also default to one hour when no timeout is specified.
Foreground calls may return a task ID after 60 seconds without resetting the command timer.
- Reduced the dependencies loaded at startup to keep the startup bundle smaller.

- Improved automatic context compaction reliability: large-context models reserve room for replies
and start compacting earlier, and adaptive-thinking models at high reasoning effort no longer fail
repeatedly when reasoning consumes the output budget.
- Update downloads follow the same proxy settings as the TUI.
- Memory tool results shown to the model are limited to 16 KiB, with a preview of the beginning and
end and guidance for reading more.
- Long sessions repeat less work when indexing and replaying history.

## 0.5.4 · 2026-09-24

### Added

- Press `Ctrl+S` in `/model` to add or remove a favorite. Favorites appear in a separate group at
the top, retain their insertion order, and persist across sessions.

### Fixed

- Scrolling back through chat remains detached from the latest output when the viewport grows or
content and the footer shrink.

## 0.5.3 · 2026-09-24

### Added

- Type `@` to select an installed Plugin. Its source identity is preserved when sending, queueing,
restoring drafts, and editing history, so plugins with the same name cannot replace one another.

### Improved

- Bash tools describe the selected shell and the current turn's background capabilities. Managed
foreground commands wait up to 60 seconds before returning a task ID, with one command deadline
covering foreground and background execution.
- Large Bash results preserve the beginning and end of the output, report process outcomes and
output persistence separately, and keep task receipts bounded.
- `/theme` uses the left and right arrow keys to select light, automatic, or dark appearance. Custom
theme watching reconciles once at startup to catch edits that could otherwise be missed.
- Workspace Skill directories (`.agents/skills`, `.claude/skills`, and `.minimax/skills`) support
symbolic links to other locations and watch linked directories for additions and changes.

### Fixed

- Pressing `Esc` before the model has produced an answer or used a tool restores the submitted
message to the composer, preserving attachments and text entered while stopping.
- Closing autocomplete, history search, settings, or multiline input restores the released chat area
correctly. Reading earlier messages keeps your position while output arrives, and sending a
message resumes following the latest output.
- Editing or rewinding history clears todos from discarded turns and stale duration indicators.
Entering the complete `/rewind` command while editing still runs the local command.
- Large-file edit previews now bound diff size and computation time. Omitted previews show a clear
notice instead of incorrectly reporting no changes.
- Goal tool calls outside a bound Goal turn return a recoverable error without ending the user turn.
- Windows managed updates validate the installed package through its Node entry point, including
installation paths containing spaces and shell metacharacters.

## 0.5.2 · 2026-09-22

### Added

- `/theme` opens a theme panel with live preview. Pick from the built-in `minimax`, `midnight`,
`graphite`, and `aurora` themes, or write your own theme file under `tui/themes/*.json`. Each
theme provides both a dark and a light palette, and the appearance follows your terminal's
background until you lock it to light or dark.
- Custom theme files support partial overrides, color aliases, syntax palettes, and hot reload, so
editing a theme you are already using takes effect without restarting MCode.
- The theme choice is stored in `tui/tui-settings.json` alongside `tuiMode` and can also be set as
`<theme-id>/light` or `<theme-id>/dark` to pin one appearance. An unrecognized value falls back to
the default theme instead of blocking startup.

### Improved

- Long conversations stay responsive by reusing verified history and loading less at runtime.
- A failed optional Provider preset sync no longer blocks startup as long as the existing
configuration stays intact.

### Fixed

- Fixed the transcript jumping back to the start of a conversation when a long turn ended after you
had scrolled up. Previously visible content now stays in place and the composer remains at the
bottom.
- Fixed style-only redraws leaving a stale composer, status bar, or duplicated body text behind in
Apple Terminal. Visible-region repaints now erase in place instead of scrolling old lines into the
terminal's native scrollback.
- Fixed failed commands being reported as successful when Bash ran through the Windows PowerShell
wrapper; the native exit code is now preserved.
- Attachment paths are resolved automatically in WSL, so Windows drive letters, UNC paths, and
`file://` URLs can be attached directly.
- Rewriting private configuration files on Windows no longer interrupts startup.
- Session history writes that wait on a busy store are retried, and stopping a turn while a write
waits keeps the turn's stopped state.

## 0.5.1 · 2026-09-21

### Improved

- Reduced repeated history processing and token estimation in long conversations.
- Provider setup now displays the Base URL and supports editing it with `ctrl+e` before testing.
Failed connection tests clearly indicate that changes were not saved. The Z.AI and Zhipu presets
label their Coding Plan and API endpoints separately.

### Fixed

- Fixed `/btw` side conversations failing to open while tools are running.
- Fixed multiline text pasted by terminals without bracketed-paste framing being submitted line by
line.
- Kept the input method candidate window anchored to the composer during redraws, including on
Windows terminals, and prevented stale history rows after content changes.
- Local sessions using custom providers or API keys no longer require a managed login for title
handling, and the welcome screen no longer asks you to sign in when the selected custom provider
does not need a MiniMax account.
- Token estimates for Z.AI and Zhipu Coding Plan endpoints no longer fall back to a local
approximation after a failed token-count request.
- Memory edits containing `$&`, `$$`, or similar sequences are inserted literally.

## 0.5.0 · 2026-09-20

### Added

- Interactive sessions support `mcode -m <provider/model>` to select a model for that session.
- Provider setup supports explicit context and output limits, image input capability
(`--support-image`), and testing before saving with `--use`.
- ACP clients can discover enabled Skills for the attached session's Agent and workspace alongside
native slash commands.
- Fullscreen Transcript supports clicking and dragging the scrollbar. An optional `context-meter`
status line item shows remaining context capacity.
- Each GitHub Release includes a versioned npm installation archive, `minimax-code-X.Y.Z.tar.gz`,
with a `.sha256` checksum.

### Improved

- Reduced repeated history processing and token estimation during long conversations.
- Provider presets import their explicit reasoning effort options.
- Source builds now use `~/.minimax` as the default data directory, the same as the published npm
CLI. Data in the earlier `~/.minimax-code` default is not moved; set `MINIMAX_DATA_DIR` to keep
using it.
- Automatic metrics and diagnostics uploads are off by default and require their own consent; global
opt-outs are honored.

### Fixed

- Preserved terminal scrollback during redraws and kept collapsed thinking summaries and the
composer visible when content shrinks.
- Settings updates now reject malformed configuration files without overwriting their contents.
- Invalid attachment markup is removed from assistant messages even when no valid attachment
remains, and path-based attachments no longer show duplicate placeholders.
- Thinking-enabled custom connections to Mistral, SiliconFlow, DashScope, and Kimi Coding send the
system prompt with a role those endpoints accept.
- Resource files of built-in and active-Agent Skills can be read in every permission mode.
- Content review failures are reported as such instead of as connectivity problems.
- Context usage is recalculated after compaction instead of showing the pre-compaction value.
- Bash deny and ask rules also match commands prefixed by transparent wrappers such as `nohup` or
`timeout`.
- Terminal control sequences in model output are sanitized before display, and reading MCode's
private runtime data requires approval.
- Configuration files that can hold credentials are restricted to owner access.
- `replace_all` edits keep literal `$` sequences.
- Session diffs report renamed files under their destination path and keep Unicode paths.
- A blank `NO_PROXY` falls back to `no_proxy`, and updates work when npm is provided through a pnpm
shell shim.

## 0.4.12 · 2026-09-18

Expand Down
Loading
Loading