cmux-next agent pane: the minimap lights where you are and the message it previews by teamleaderleo · Pull Request #22879 · manaflow-ai/cmux

teamleaderleo · 2026-10-21T03:33:56Z

The thread minimap now shows where you are and which message it is previewing, as ChatGPT's rail does. It also hides on chats that don't scroll. This follows up #22784 with Leo's feedback and #22784's review.

Changes

Previewed message stays lit (both rails). While the pointer or keyboard focus is on a prompt's preview card, that prompt's tick stays highlighted. Before, moving from the rail onto the card set the tick back to rest, so the card pointed at nothing.
Where you are (full map). Marks under the view bar draw as current, brighter than the rest, at the top and at the bottom of the chat, and they stay that way while the pointer is on the rail. The covered marks are recomputed in the bar's own scroll handler, and React renders only when the covered set changes.
No map for a chat that doesn't scroll. ScrollMap compared the layout, which includes the composer's dock space, with the viewport above the dock. A chat slightly shorter than its pane drew a map whose bar filled the rail. The transcript now passes scrolls, using the same test as its bottom fade.
Quiet card during a scrub. Dragging dismisses the preview card and no longer reopens it on every mark it crosses. Hover after the drag opens it again.

Tests

test/thread-scroll-map-rail.test.tsx covers three things: current marks at the top and the bottom, the previewed prompt staying lit while focus is on its card's bookmark, and no map on a chat that fits its pane.
test/thread-bookmarks.test.tsx passes the new scrolls prop.

Observed / Not run

Observed: scripts/web-lint.sh and prepush-gate.py passed.
Not run locally: the tests (no test runs on Leo's Mac). CI is the first run.
No test covers the dock-space case of the "doesn't scroll" fix, because jsdom does not measure the dock. The no-map test covers a plainly short chat.
Leo's screenshot showed the compact rail, which the build he was running still has. The full map from cmux-next agent pane: the thread minimap spans the transcript's height #22784 is the default now. The "stays lit" fix covers both rails.

Changelog

Changed: The chat's minimap brightens the messages on screen and keeps the previewed message highlighted while you read its preview.

Agent session: cosmic-gray

🤖 Generated with Claude Code

^{Need help on this PR? Tag @codesmith-bot with what you need.}

Summary by cubic

The thread minimap now lights the messages on screen in your current view and keeps the previewed message highlighted while you read its preview, and it no longer appears on chats that don't scroll.

Marks under the view bar render as current, brighter than the rest, at the top and bottom of the chat; bookmarked ticks among them keep full opacity while hovering the rail.
The previewed prompt's tick stays lit while the pointer or keyboard focus is on its card.
A chat that fits its pane above the composer no longer draws a map; the transcript now passes scrolls, the same test its bottom fade uses, where ScrollMap previously compared the full layout with the viewport.
Scrubbing the rail dismisses the preview card and doesn't reopen it on every mark crossed; hovering after the drag opens it again.

^{Written for commit bab3c97. Summary will update on new commits.}

…ll, and holds its card during a scrub - ScrollMap compared the layout (which counts the composer's dock space) with the viewport above the dock, so a chat a little shorter than its pane showed a map whose bar filled the rail. The transcript now says whether it scrolls, the way its bottom fade already decides. - A scrub dismisses the preview card and no longer reopens it on each mark crossed; hovering after the drag opens it again. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a15ba9-9614-7832-8e39-d8ca0407c193&host=cosmic-gray

… previews Leo: the hover preview did not highlight its message, and the map should show where you are, at the bottom too (ChatGPT's rail does both). - Both rails: the previewed prompt's tick stays lit while the pointer or focus is on its card. Before, leaving the rail for the card put it back to rest. - Full map: the marks under the view bar draw as current, brighter than the rest, and stay so while the pointer is on the rail. The bar's covered marks are recomputed on its own scroll frame and render only when they change. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a15ba9-9614-7832-8e39-d8ca0407c193&host=cosmic-gray

teamleaderleo · 2026-10-21T03:41:16Z

Bead cx-cjt8o created for this PR by the cmux coordinator.

github-actions · 2026-10-21T03:42:40Z

cmux status _{(one comment, edited in place; each line is updated by the job that owns it)}

Unclear: web / react-apps-check (3) failed: no known signature; failed step: Build the web bundles. This PR merged before its CI finished: fix forward on main.

CI failed on bab3c97db0 (run 39540753617 attempt 1): 5 unknown.

Job	Verdict	Why
web / react-apps-check (3)	unknown	no known signature; failed step: `Build the web bundles`
web / react-apps-check (2)	unknown	no known signature; failed step: `Build the web bundles`
web / react-apps-check (1)	unknown	no known signature; failed step: `Build the web bundles`
web / react-apps-check (4)	unknown	no known signature; failed step: `Build the web bundles`
web / react-apps-check	unknown	no known signature; failed step: `Check webview test shards`

Not re-run automatically: web / react-apps-check (3), web / react-apps-check (2), web / react-apps-check (1), web / react-apps-check (4), web / react-apps-check are not machine failures.

Written by scripts/ci/classify_failures.py (ci-failure-attribution.yml); signatures are its SIGNATURES table. A machine verdict is the runner's fault, not this PR's; yours means the failing file is one this PR changes, also red on main that main's latest full suite fails the same way, seen on other PRs that it failed on another pull request's run lately.

red: web / Web status; base red: the same failure on the base dea0d5a; does not block
red: web / react-apps-check; base red: the same failure on the base dea0d5a; does not block
red: web / react-apps-check (1); base red: the same failure on the base dea0d5a; does not block
red: web / react-apps-check (2); base red: the same failure on the base dea0d5a; does not block
red: web / react-apps-check (3); base red: the same failure on the base dea0d5a; does not block
red: web / react-apps-check (4); base red: the same failure on the base dea0d5a; does not block

Tour of bab3c97db0 (run): no media. The tour did not run: only webview code changed (6 path(s)), which the gallery advisory run checks in Linux Chromium on the pull request (its artifact holds any captures); add the media-tour-mac label to tour it on a Mac.

…ity while hovering the map Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a15ba9-9614-7832-8e39-d8ca0407c193&host=cosmic-gray

teamleaderleo · 2026-10-21T03:52:24Z

owner: 01a15ba9-9614-7832-8e39-d8ca0407c193

teamleaderleo · 2026-10-21T03:55:51Z

Merging under the base-red rule. Every red here is red on base too:

react-apps-check fails at "Build the web bundles" on a stale Native/OptChat/optchat-chief/inspector/index.html.
defaults-snapshot.
The review-link job.

Not run: the base bundle red stops this PR's webview tests (jsdom and browser) before they run, so the new tests are unverified in CI. The independent review passed (code read only).

teamleaderleo · 2026-10-21T03:56:35Z

high-signal review link / link is red on 20 of 26 fresh feat-cmux-next PR heads in the last 2:00:00, so it is most likely a base red, not this PR's. The base-reds lane has it; a base merge picks up the fix once it lands.

github-actions · 2026-10-21T08:10:22Z

Shipped in:

cmux NEXT nightly 1.0.0-nightly.3955987471201 (03960f5a64)

github-actions · 2026-10-21T08:25:57Z

Shipped in:

cmux NEXT nightly 1.0.0-nightly.3956132659001 (233cc81754)

…23285) check-scrollbars (SCROLLBARS-THIN) fails cmux-next checks on feat-cmux-next: the thread minimap's list (#22879) and the Apps Discover chips hid their scrollbars with scrollbar-width and ::-webkit-scrollbar, which turn WebKit's overlay scroller into a custom one. They now keep the shared thin overlay scroller, hidden until the user scrolls. That takes the check from 13 hits to 10, back under its ceiling. Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a15d18-aabc-7850-af29-f0655b1d2a1a&host=cosmic-gray Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

…e left edge) (#23735) * test(webviews): the agent chat ships with no thread minimap (red) The default arm of thread-minimap-layout is full since #22784, so every agent chat draws the full-height ScrollMap (a dash per message and a view box) down the transcript's left edge, and the compact rail's list shows a second scroller since #23285. These tests fail until the default is off and the list stops scrolling on its own. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a16027-a80b-79c0-ab31-500e58883d27&host=cosmic-gray * fix(webviews): the agent chat draws no thread minimap by default Leo on the nightly (2026-10-21): a column of dashes and a thumb box down the left edge of the chat, and a second scroller. That is the full-height ScrollMap, the default arm since #22784 (lit band #22879), and the compact rail's own list, which shows a scroller since #23285 dropped its hiding. - thread-minimap-layout gets an off arm, and off is the default: no map, no rail, nothing at the left edge. Previous and Next Turn and the bookmarked-prompt steps still work. full and compact stay selectable from cmux.experiments for dogfood. - The compact rail's list is overflow: hidden: it still scrolls from script and focus (band follow, scrub, Tab) but never draws a scroller, so the transcript keeps its one shared thin scroller (SCROLLBARS-THIN). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a16027-a80b-79c0-ab31-500e58883d27&host=cosmic-gray * style(webviews): format the minimap experiment and its test Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

teamleaderleo and others added 2 commits October 20, 2026 20:29

teamleaderleo added the risk:normal Risk tier (docs/ci/risk-tiers.md) label Oct 21, 2026

This was referenced Oct 21, 2026

Agent chat: a steer is one continuous turn; only Stop or Esc reads as stopped #22691

Merged

cmux-next: thread minimap matches the Codex rail; bookmarks show and jump #20560

Closed

fix(agent-pane): a bookmarked mark under the view bar keeps full opac…

bab3c97

…ity while hovering the map Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Agent-Session: https://cmux.com/deeplink/agent?harness=claude&session=01a15ba9-9614-7832-8e39-d8ca0407c193&host=cosmic-gray

teamleaderleo marked this pull request as ready for review October 21, 2026 03:55

teamleaderleo merged commit 33596f5 into feat-cmux-next Oct 21, 2026

teamleaderleo deleted the cmux/minimap-followups branch October 21, 2026 03:55

This was referenced Oct 21, 2026

ci: every build-fleet client call checks the client is installed #23248

Merged

cmux NEXT: minimap rail and Apps chips keep the shared scroller (fixes the check-scrollbars base red) #23285

Merged

teamleaderleo mentioned this pull request Oct 21, 2026

Agent chat: no thread minimap by default (the column of marks down the left edge) #23735

Merged

Navigation Menu

Repository navigation

cmux-next agent pane: the minimap lights where you are and the message it previews - #22879

cmux-next agent pane: the minimap lights where you are and the message it previews#22879
teamleaderleo merged 3 commits into
feat-cmux-nextfrom
cmux/minimap-followups

teamleaderleo commented Oct 21, 2026 •

edited by cubic-dev-ai Bot

Loading

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026 •

edited

Loading

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

teamleaderleo commented Oct 21, 2026 • edited by cubic-dev-ai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes

Tests

Observed / Not run

Changelog

Summary by cubic

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

teamleaderleo commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

github-actions Bot commented Oct 21, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

teamleaderleo commented Oct 21, 2026 •

edited by cubic-dev-ai Bot

Loading

github-actions Bot commented Oct 21, 2026 •

edited

Loading