Truncating the transcript is the mechanical fix, but the turns being cut are the
work itself. Losing the conversation that produced a morning's changes is most
of the damage, and a session that resumes with a hole in its memory is barely
resumed - it has no idea what it just did or what it was asked.
So --apply now writes the dropped turns out as <session>.recovered-<stamp>.md,
a readable record of what was asked, what was answered and what was run, and
appends the same text back to the truncated transcript as a single note. Tool
results are deliberately excluded: they are most of a transcript by volume and
the least useful part of a summary.
The note is a user entry marked isMeta - the marker Claude Code already uses for
its own local-command caveats, meaning context rather than something to answer -
and carries no message.id, so it cannot recreate the previous_message_id
condition being repaired. Verified on the real corrupted transcript: 1921 lines
in, 1813 kept plus the note, chained to the last good assistant uuid, no id on
the message, and the file reads back as healthy. The generated digest recovers
the actual instruction that was lost in the incident, which is the thing that
made this worth doing.
--no-reinject writes the markdown but leaves the session alone.
Acts on docs/incident-mode-switch-corrupts-live-sessions.md, which is added here
as the record of why.
The report identifies a consequence that was not modelled. A failed call is
recoverable; a *successful* one may not be. If a running session takes even one
completion from the provider being switched to - which happens when that mode
matches the base URL it already had cached - that provider's message-id format
lands in its transcript. OpenRouter issues `gen-<epoch>-<rand>` where Anthropic
issues `msg_...`, and native Anthropic then refuses to resume the session at
all, with a 400 naming previous_message_id. The only way back is to truncate the
transcript, losing every turn after the cut. That happened here, and was fixed
by hand.
Two changes follow.
Sessions are now settled before the write, not reported after it. A switch with
anything running stops, names the sessions, explains what is about to happen to
them, and offers restart (the only answer that ends with everything on the mode
the bar now claims), close, proceed anyway, or abort - defaulting to abort.
Non-interactively it refuses outright unless given --yes. The old after-the-fact
reporter is deleted rather than left as a second, contradictory account.
`claude-mode repair-session` replaces the hand surgery: it finds a project's
transcripts, reports which are resumable, and on --apply backs the file up and
truncates to the last Anthropic-issued message. Verified against the real
corrupted transcript from the incident - it reproduces the manual cut exactly,
1921 lines to 1813, dropping the two `gen-` completions and the error
placeholders after them, leaving a transcript that ends on a genuine msg_ id. It
refuses a transcript written to in the last 90 seconds, since that one belongs
to a session still running.
The panel passes --yes, having already asked in its own card, and that card now
names the transcript risk rather than only the inconvenient one.
Requirement 4 of the report - documenting the mechanism - landed in 9c301e1;
the README now carries the unrecoverable half as well.
The openrouter default moves its hot tiers: opus to z-ai/glm-5.3-flash and
sonnet to deepseek/deepseek-v4-flash-0731. haiku and fable are unchanged.
`cheap` and `lmstudio-qwen` are gone, leaving exactly one preset per mode so
`claude-mode <mode>` is never ambiguous and there is no menu to read before the
thing you asked for happens. The surviving lmstudio preset keeps the Qwen3.6
model rather than KAT-Coder: the two differed mainly in that KAT's chat template
carries the message-order assertion this README already warns about, so between
two presets that had to become one, the one that is known to work won.
More presets are still a `preset new` away; the shipped set is a starting point,
not a ceiling.
Which is the other half of this. A shipped preset was never a working
configuration - OpenRouter and Z.AI have no key stored, and lmstudio's model ids
were whatever happened to be installed on the machine this was packaged on. That
was left for the user to discover through a failure. Now the shipped presets
carry `configured: false`, preflight blocks on it, and `claude-mode setup <mode>`
walks through what is actually needed: key, server URL and auth for LM Studio,
then models chosen from the provider's own catalogue rather than typed from
memory. A switch that trips this in a terminal offers to run setup there and
then instead of printing a command to type next.
Absent means configured, deliberately: presets that predate this and any built
by hand with `preset new` do not suddenly start demanding a wizard.
The panel gets a "Set up <mode>…" button that hands the whole flow to a terminal,
since a bar popup can host neither a hidden key prompt nor a filter-select list.
Two bugs found while testing it, both real:
ask_value printed its prompt to stdout while being called inside $( ), so the
prompt text came back glued to the front of the answer and set-url rejected the
result. Moved to stderr, which is why warn and err already go there.
lms_catalogue never sent the API key. On a server with authentication switched
on - the case just added support for - /api/v0/models answers 401 like anything
else, so the catalogue came back empty and every caller silently concluded the
server had no models installed. It now sends the preset's credential, as do the
three other call sites that read it.
Every one of these said a switch leaves running sessions alone - that they keep
talking to the old provider until restarted. That is wrong, and reassuring in
exactly the wrong direction: a switch breaks them.
The static half of the config does behave as described. Base URL, model ids and
the env block are read once at startup and a running session keeps what it
started with.
The credential is not. It comes from running apiKeyHelper, which Claude Code
re-invokes on a timer - CLAUDE_CODE_API_KEY_HELPER_TTL_MS is present in the
2.1.251 binary - and claude-key-helper.sh answers for whatever state.json says at
that moment. So the switch reaches into a live session through the one part that
was never cached. Into anthropic, the helper returns nothing by design and the
next refresh comes back with no credential; into another provider, it hands over
the new key while the session is still pointed at the old base URL, which
rejects it. Either way the session starts failing calls whenever the TTL happens
to expire, mid-turn as easily as between turns.
One case does survive, and is now stated rather than glossed: switching between
two presets of the same provider that share a keyRef keeps the same key and the
same endpoint, so the session carries on with the model ids it started with.
The claim in "Restarting sessions" predates this work - it came in with the
Windows build - but it is wrong for the same reason and is corrected too.
GLM 5.3 supersedes 5.2, so every reference to the older id follows it: the zai
preset's opus, sonnet and fable tiers, and the fable tier of `default`, which
routes to the same model through OpenRouter.
Also the hardcoded Z.AI entries in both builds' model pickers, which are the one
place a model id is not read from a preset - leaving those would have gone on
offering a superseded model from the menu while the presets used the new one.
haiku stays on glm-4.7: that is the fast/cheap tier, not a 5.x.
A Quickshell plugin for the Omarchy shell (v4's bar is Quickshell, not Waybar).
The icon is the mode; clicking it opens a panel that switches.
State comes from watching health.json rather than polling the CLI, so the widget
costs nothing while idle and a switch made in a terminal shows up in the bar on
its own. state.json is the fallback, because it exists from install onward where
health.json only appears once claude-mode has run.
The marks are the providers' real logos, drawn as vector paths through
QtQuick.Shapes rather than set as font glyphs - three of the four have no Nerd
Font pictograph at all, and paths take the bar's foreground colour and follow
the theme. Claude, OpenRouter and LM Studio from simple-icons, Z.AI from
lobe-icons; trademarks belong to their owners. Zhipu's mark was rejected for
Z.AI: it is a dense lattice that is unreadable below ~40px.
Rendering them at 13px took three fixes. A layer rasterises the Shape at its own
size and then scales the texture, so a 24px buffer minified to 13 resampled two
pixels into one and the Claude burst lost rays; without the layer the scale is a
transform on the geometry and rasterisation happens once, at final resolution.
CurveRenderer replaces the tessellating default. And the size is forced odd,
because a radially symmetric mark puts its vertical and horizontal arms on the
centre line, which is a pixel centre at odd sizes and the seam between two
pixels at even ones, where each arm splits its coverage and greys out. Measured
against a cairo render at the same size, the result is now identical.
Sizing is measured rather than guessed. Every stock glyph in this bar paints
11px of ink; the marks fill their box instead of carrying a font's padding, so
the box is the smaller number. Each mark also carries an optical scale from two
measurements of a 200px render - LM Studio's filled container covers 69% of its
box against ~38% for the others, and Z.AI and OpenRouter are wide-but-short
marks spanning ~84% of the box height.
Choosing a target does not switch immediately. It runs the CLI's own preflight,
and if that refuses, the panel names what is missing and offers the fix - a
terminal for the hidden key prompt, a re-check once a server is up, or the
server form. If sessions are running it lists them by terminal and directory,
marks any mid-request, and asks whether to restart them, close them, or leave
them. The session action is applied strictly after the write, since restarting
first would only bring them back up on the provider just left.
The server form edits an LM Studio preset's base URL and whether it needs an API
key, reachable from the gear on any LM Studio preset row - per row, because two
presets can point at two different machines - and from the failure card. The
shipped local default is untouched unless it is changed.
Source of truth so far has been c:\Users\smoido\projects\cli on the Windows
box, which has no git history of its own. This is that tree copied verbatim over
SSH, minus dist/ - the PowerShell build, the POSIX port under linux/, and the
presets both share.
Recorded as its own commit so that everything after it is a reviewable diff
rather than an undifferentiated first drop.
smoido